Tag Info

New answers tagged

0

There are a number of open source solutions like those mentioned above (munin, zabbix, nagios, OpenTSDB). There are also commercial solutions like Metrink, DataDog, and New Relic. You'll really want to find something that can monitor more than just CPU and disk, as the problem can be in any number of places.


2

As with any performance tuning there are no absolute rules. Due to the nature of it, there are a few rules that you can make with sufficient qualification but only a few. So bear that in mind. How you should control for hardware interrupts really depends on how your workload behaves For your question, you also have to control where the interrupts are sent. ...


5

When an server process starts it issues some system calls (socket() and listen()). The system then opens the port and creates a socket file descriptor for the process to interact with. You can see this with: Find the Apache master process id: root@frisbee:~# ps -ef | grep apache | grep root root 27440 1 0 16:06 ? 00:00:00 /usr/sbin/apache2 ...


0

You problem is probably because you are reading /proc/cputime from two separate processes. The idle time will will increase slightly between each cat, giving the possibility for a lower reading the second time. I recommend doing this instead: read uptime idle </proc/cputime echo -e "scale=10\n ($uptime-($idle/16))*100" | bc Also, if you want your ...


4

(uptime-(idle_time/num_core)) May give an idea of how long the system has been busy, in seconds. Multipling that by 100 makes it centiseconds -- is that your intention? IMO it would make more sense to consider how many processor seconds in total were available, and subtract the idle time from that: uptime * num_core - idle_time = total active ...


1

I have to use two dashes for this parameter, like $ ps --ppid 1 My version: $ ps --version procps-ng version 3.3.4


0

I generally switch to another virtual console. I'm using Fedora 19 so Ctrl+Alt+F1 would be the primary one you're using now to display X. So I'll often times switch to Ctrl+Alt+F2 where I can either maintain a logged in console, or I can quickly login to there and run the necessary kill commands to halt whatever process is going wild. Other distros make ...


2

If you run a fake job such as sleep 120 and then watch it in htop you'll notice that its state is S aka "SLEEP" and the processes TIME remains at 0:00.00 for the duration. That's because that process is consuming 0 CPU time, which is the intent of the TIME column. It tracks the amount of CPU time a given process has used. ...


2

Your CPU is a dual core CPU with hyperthreading Intel® Core™ i5-460M Processor This means you have 2 cores and they are physical CPU's. You have also hyperthreading and so you have 4 logical CPU's. taskset was designed because the balancing of tasks in a multicore CPU was a performance lost, they did normally not use hyperthreading. You have a ...


1

LUKS is a block device encryption layer which sits on top of a block device and encrypts/decrypts all accesses to that device. No unencrypted data ever touches the physical device. LUKS then provides a virtual block device which gets used by the system to access the files. It is thus transparent to applications, which have no idea that encryption is taking ...


6

Here is a script to print the total CPU usage for each user currently logged in, showPerUserCPU.sh: own=$(id -nu) for user in $(who | awk '{print $1}' | sort -u) do # print other user's CPU usage in parallel but skip own one because # spawning many processes will increase our CPU usage significantly if [ "$user" = "$own" ]; then continue; fi ...



Top 50 recent answers are included