2

I've got an old MacBook Pro with a Core2Duo CPU that regularly has a load up in the 3s or 4s.

thumper:~ joliver.sa$ sysctl -n hw.ncpu
2
thumper:~ joliver.sa$ uptime
 9:24  up 4 days, 16:31, 3 users, load averages: 3.50 4.09 3.83

Processes: 295 total, 2 running, 14 stuck, 279 sleeping, 1225 threads                                                                                                                                  09:24:43
Load Avg: 3.59, 4.06, 3.83  CPU usage: 4.83% user, 6.28% sys, 88.88% idle  SharedLibs: 17M resident, 20M data, 0B linkedit. MemRegions: 48719 total, 3030M resident, 159M private, 655M shared.
PhysMem: 7200M used (1446M wired), 991M unused. VM: 698G vsize, 1071M framework vsize, 0(0) swapins, 0(0) swapouts. Networks: packets: 38837251/53G in, 7236342/639M out.
Disks: 1045731/15G read, 1170955/228G written.

This clearly isn't a CPU issue. It doesn't appear to be a memory issue. Disk IO doesn't seem to be an issue. Some of the "stuck" processes were things I could kill off, but the load has only gotten worse in the last few minutes. How else can I find out what's lining up waiting for attention?

Improve this question
5
  • any highly repetitive messages in the Console ?
    – Ruskes
    Apr 21, 2015 at 16:58
  • No... system logs are pretty quiet. Nothing that looks out of place.
    – John Oliver
    Apr 21, 2015 at 17:02
  • I was comparing your (TOP) results Load Avg: 0.79, 0.77, 0.83 CPU usage: 2.6% user, 3.66% sys, 94.26% idle.....and the rest look similar, other than your Network is way higher.
    – Ruskes
    Apr 21, 2015 at 17:18
  • Did you run the "ps aux" and checked there what is going on!
    – Ruskes
    Apr 21, 2015 at 17:22
  • I have, of course, used ps. There are some Z or U processes, but other Macs with no load problems have just as many or more. And there's very little network traffic.
    – John Oliver
    Apr 21, 2015 at 17:30

1 Answer 1

Reset to default
1

Here are some troubleshooting steps:

Run ps aux to see which ones are actually running real time.

Next run the iostat to watch the actual CPU loading and see if you can determine if some new processes are running and whether that is increasing or decreasing the IO and CPU load on the system.

For comparison here is the output of the iostat I have.

disk0           disk1           disk2       cpu     load average
    KB/t tps  MB/s     KB/t tps  MB/s     KB/t tps  MB/s  us sy id   1m   5m   15m
   26.55   3  0.08   177.96   0  0.00    65.01   0  0.00   3  2 96  1.16 0.96 0.90

Some Git and Homebrew are know to create the problem so check your versions to latest update.

Improve this answer
4
  • There is very little running that isn't sleeping (can't paste output, too many characters, but not even 20 processes) And I'm losing formatting for iostat output so I'll leave one line for readability, they're all very similar: thumper:~ joliver.sa$ iostat 1 disk0 cpu load average KB/t tps MB/s us sy id 1m 5m 15m 0.00 0 0.00 3 2 95 4.70 5.38 5.53 I was going to say there's no git or homebrew on this Mac, but there is a /usr/bin/git If this was Linux, I could quickly find out what it belongs to, but on a Mac... I don't know.
    – John Oliver
    Apr 21, 2015 at 17:57
  • are you using Apple git and/or Homebrew git ? if yes what are the versions.
    – Ruskes
    Apr 21, 2015 at 18:03
  • No. It isn't impossible they might be on there somewhere, but how would I know? I'm assuming /usr/bin/git is from Apple. I have no idea what those packages would look like, where they'd install, etc.
    – John Oliver
    Apr 21, 2015 at 18:56
  • Check this out github.com/robbyrussell/oh-my-zsh/issues/2189
    – Ruskes
    Apr 21, 2015 at 19:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .