Hack and / - Linux Troubleshooting, Part I: High Load
Listing 1. Sample top Output
top - 14:08:25 up 38 days, 8:02, 1 user, load average: 1.70, 1.77, 1.68 Tasks: 107 total, 3 running, 104 sleeping, 0 stopped, 0 zombie Cpu(s): 11.4%us, 29.6%sy, 0.0%ni, 58.3%id, .7%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 1024176k total, 997408k used, 26768k free, 85520k buffers Swap: 1004052k total, 4360k used, 999692k free, 286040k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 9463 mysql 16 0 686m 111m 3328 S 53 5.5 569:17.64 mysqld 18749 nagios 16 0 140m 134m 1868 S 12 6.6 1345:01 nagios2db_status 24636 nagios 17 0 34660 10m 712 S 8 0.5 1195:15 nagios 22442 nagios 24 0 6048 2024 1452 S 8 0.1 0:00.04 check_time.pl
As you can see, there's a lot of information in only a few lines. The first line mirrors the information you would get from the uptime command and will update every few seconds with the latest load averages. In this case, you can see my system is busy, but not what I would call heavily loaded. All the same, this output breaks down well into our different load categories. When I troubleshoot a sluggish system, I generally will rule out CPU-bound load, then RAM issues, then finally I/O issues in that order, so let's start with CPU-bound load.
CPU-bound load is load caused when you have too many CPU-intensive processes running at once. Because each process needs CPU resources, they all must wait their turn. To check whether load is CPU-bound, check the CPU line in the top output:
Cpu(s): 11.4%us, 29.6%sy, 0.0%ni, 58.3%id, .7%wa, 0.0%hi, 0.0%si, 0.0%st
Each of these percentages are a percentage of the CPU time tied up doing a particular task. Again, you could spend an entire column on all of the output from top, so here's a few of these values and how to read them:
us: user CPU time. More often than not, when you have CPU-bound load, it's due to a process run by a user on the system, such as Apache, MySQL or maybe a shell script. If this percentage is high, a user process such as those is a likely cause of the load.
sy: system CPU time. The system CPU time is the percentage of the CPU tied up by kernel and other system processes. CPU-bound load should manifest either as a high percentage of user or high system CPU time.
id: CPU idle time. This is the percentage of the time that the CPU spends idle. The higher the number here the better! In fact, if you see really high CPU idle time, it's a good indication that any high load is not CPU-bound.
wa: I/O wait. The I/O wait value tells the percentage of time the CPU is spending waiting on I/O (typically disk I/O). If you have high load and this value is high, it's likely the load is not CPU-bound but is due to either RAM issues or high disk I/O.
If you do see a high percentage in the user or system columns, there's a good chance your load is CPU-bound. To track down the root cause, skip down a few lines to where top displays a list of current processes running on the system. By default, top will sort these based on the percentage of CPU used with the processes using the most on top (Listing 2).
Listing 2. Current Processes Example
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 9463 mysql 16 0 686m 111m 3328 S 53 5.5 569:17.64 mysqld 18749 nagios 16 0 140m 134m 1868 S 12 6.6 1345:01 nagios2db_status 24636 nagios 17 0 34660 10m 712 S 8 0.5 1195:15 nagios 22442 nagios 24 0 6048 2024 1452 S 8 0.1 0:00.04 check_time.pl
The %CPU column tells you just how much CPU each process is taking up. In this case, you can see that MySQL is taking up 53% of my CPU. As you look at this output during CPU-bound load, you probably will see one of two things: either you will have a single process tying up 99% of your CPU, or you will see a number of smaller processes all fighting for a percentage of CPU time. In either case, it's relatively simple to see the processes that are causing the problem. There's one final note I want to add on CPU-bound load: I've seen systems get incredibly high load simply because a multithreaded program spawned a huge number of threads on a system without many CPUs. If you spawn 20 threads on a single-CPU system, you might see a high load average, even though there are no particular processes that seem to tie up CPU time.
Kyle Rankin is a VP of engineering operations at Final, Inc., the author of a number of books including DevOps Troubleshooting and The Official Ubuntu Server Book, and is a columnist for Linux Journal. Follow him @kylerankin.
Practical books for the most technical people on the planet. Newly available books include:
- Agile Product Development by Ted Schmidt
- Improve Business Processes with an Enterprise Job Scheduler by Mike Diehl
- Finding Your Way: Mapping Your Network to Improve Manageability by Bill Childers
- DIY Commerce Site by Reven Lerner
Plus many more.
- Non-Linux FOSS: Snk
- diff -u: What's New in Kernel Development
- Server Hardening
- 22 Years of Linux Journal on One DVD - Now Available
- Building a Multisourced Infrastructure Using OpenVPN
- Giving Silos Their Due
- Controversy at the Linux Foundation
- Don't Burn Your Android Yet
- What's New in 3D Printing, Part III: the Software