Grokbase › Groups › Node.js › nodejs › November 2012
FAQ
We're having a problem where every once in a while one of our environments
our node app runs on 100% CPU. The server isn't very active and usually
runs on 0%-2% CPU.
I was wondering what are the common issues that might cause this problem
and what would be the best way to find out what causing this issue.

technical spec:
node version 0.8.14
ubuntu 11.10
Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

gems used:
"express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
"0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
"0.x.x",

--
Job Board: http://jobs.nodejs.org/
Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
nodejs+[email protected]
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

Search Discussions

  • Chad Engler at Nov 14, 2012 at 3:49 pm ⇧
    Hehe, gems...



    Sounds like garbage collection. Do you create large objects or buffers
    temporarily?



    -Chad



    From: [email protected] On Behalf
    Of Dor Tzur
    Sent: Wednesday, November 14, 2012 4:08 AM
    To: [email protected]
    Subject: [nodejs] nodejs 100% cpu randomly



    We're having a problem where every once in a while one of our
    environments our node app runs on 100% CPU. The server isn't very active
    and usually runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local"
    : "0.x.x",



    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Anand George at Nov 14, 2012 at 4:01 pm ⇧
    Are you using a service like forever? Had an API server running using
    forever. Running without forever showed a CPU usage of about 5%. Running
    with forever, the CPU usage would go to 100% on the first few calls. Using
    another tool solved the issue.
    On Wed, Nov 14, 2012 at 9:19 PM, Chad Engler wrote:

    Hehe, gems…****

    ** **

    Sounds like garbage collection. Do you create large objects or buffers
    temporarily?****

    ** **

    -Chad****

    ** **

    *From:* [email protected] *On
    Behalf Of *Dor Tzur
    *Sent:* Wednesday, November 14, 2012 4:08 AM
    *To:* [email protected]
    *Subject:* [nodejs] nodejs 100% cpu randomly****

    ** **

    We're having a problem where every once in a while one of our environments
    our node app runs on 100% CPU. The server isn't very active and usually
    runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",****

    ** **

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en****

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Adam Crabtree at Nov 14, 2012 at 4:40 pm ⇧
    On the note of forever running at 100%, that likes to happen when your
    process fails to start and forever attempts to restart it as fast as it can
    die.

    On Wed, Nov 14, 2012 at 8:01 AM, Anand George wrote:

    Are you using a service like forever? Had an API server running using
    forever. Running without forever showed a CPU usage of about 5%. Running
    with forever, the CPU usage would go to 100% on the first few calls. Using
    another tool solved the issue.

    On Wed, Nov 14, 2012 at 9:19 PM, Chad Engler wrote:

    Hehe, gems…****

    ** **

    Sounds like garbage collection. Do you create large objects or buffers
    temporarily?****

    ** **

    -Chad****

    ** **

    *From:* [email protected] *On
    Behalf Of *Dor Tzur
    *Sent:* Wednesday, November 14, 2012 4:08 AM
    *To:* [email protected]
    *Subject:* [nodejs] nodejs 100% cpu randomly****

    ** **

    We're having a problem where every once in a while one of our
    environments our node app runs on 100% CPU. The server isn't very active
    and usually runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",****

    ** **

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en****

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en


    --
    Better a little with righteousness
    than much gain with injustice.
    Proverbs 16:8

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Anand George at Nov 14, 2012 at 4:50 pm ⇧
    The app has no issues running without forever... so not sure why forever
    wants to restart the process. The app has been running for about 6 weeks
    now using nohup with supervisor on AWS, have had at least six code changes
    and have never faced this issue.
    On Wed, Nov 14, 2012 at 10:10 PM, Adam Crabtree wrote:

    On the note of forever running at 100%, that likes to happen when your
    process fails to start and forever attempts to restart it as fast as it can
    die.

    On Wed, Nov 14, 2012 at 8:01 AM, Anand George wrote:

    Are you using a service like forever? Had an API server running using
    forever. Running without forever showed a CPU usage of about 5%. Running
    with forever, the CPU usage would go to 100% on the first few calls. Using
    another tool solved the issue.

    On Wed, Nov 14, 2012 at 9:19 PM, Chad Engler wrote:

    Hehe, gems…****

    ** **

    Sounds like garbage collection. Do you create large objects or buffers
    temporarily?****

    ** **

    -Chad****

    ** **

    *From:* [email protected] *On
    Behalf Of *Dor Tzur
    *Sent:* Wednesday, November 14, 2012 4:08 AM
    *To:* [email protected]
    *Subject:* [nodejs] nodejs 100% cpu randomly****

    ** **

    We're having a problem where every once in a while one of our
    environments our node app runs on 100% CPU. The server isn't very active
    and usually runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",****

    ** **

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en****

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en


    --
    Better a little with righteousness
    than much gain with injustice.
    Proverbs 16:8

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Michael Schoonmaker at Nov 14, 2012 at 5:45 pm ⇧
    ... which is easily diagnosed with `forever list`, and can be assuaged
    with forever's "spinning" parameters, "minUptime" and "spinSleepTime".

    Other than Adam's reason, I've never seen CPU spikes because of
    forever. Let's not have any straw men here.
    On Wed, Nov 14, 2012 at 8:40 AM, Adam Crabtree wrote:
    On the note of forever running at 100%, that likes to happen when your
    process fails to start and forever attempts to restart it as fast as it can
    die.

    On Wed, Nov 14, 2012 at 8:01 AM, Anand George wrote:

    Are you using a service like forever? Had an API server running using
    forever. Running without forever showed a CPU usage of about 5%. Running
    with forever, the CPU usage would go to 100% on the first few calls. Using
    another tool solved the issue.


    On Wed, Nov 14, 2012 at 9:19 PM, Chad Engler <[email protected]>
    wrote:
    Hehe, gems…



    Sounds like garbage collection. Do you create large objects or buffers
    temporarily?



    -Chad



    From: [email protected] On Behalf
    Of Dor Tzur
    Sent: Wednesday, November 14, 2012 4:08 AM
    To: [email protected]
    Subject: [nodejs] nodejs 100% cpu randomly



    We're having a problem where every once in a while one of our
    environments our node app runs on 100% CPU. The server isn't very active and
    usually runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",



    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en



    --
    Better a little with righteousness
    than much gain with injustice.
    Proverbs 16:8

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Jimb Esser at Nov 14, 2012 at 5:50 pm ⇧
    If it spikes to 100%, and then goes back down, I'd agree that sounds like
    garbage collection.

    Either way, the best way to know what is going on is to profile it! I
    think on current versions of node, node-webkit-agent[1] is currently the
    best (only?) way to get a JavaScript CPU profile. If it reports all of the
    time in "(program)", then you may need a lower-level CPU profile to know
    what's going on.

    [1] https://github.com/c4milo/node-webkit-agent

    Jimb Esser
    On Wednesday, November 14, 2012 1:08:04 AM UTC-8, Dor Tzur wrote:

    We're having a problem where every once in a while one of our environments
    our node app runs on 100% CPU. The server isn't very active and usually
    runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Dor Tzur at Nov 15, 2012 at 8:41 am ⇧
    A. We are not using forever so it's not relevant to this case, thanks for
    the input though.

    B. We tried using nodetime but it couldn't profile the cpu when it's at
    100%. Do you think node-web-kit has a better chance?

    C. We've been using MongoDB without defining safe mode and with
    auto-reconnect turned on. do you think one of these options might cause
    this?
    On Wednesday, November 14, 2012 11:08:04 AM UTC+2, Dor Tzur wrote:

    We're having a problem where every once in a while one of our environments
    our node app runs on 100% CPU. The server isn't very active and usually
    runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Anton Whalley at Nov 15, 2012 at 9:41 am ⇧
    Can you put it on an illumOS based system temporarily and get DTracing?
    At the very least this would tell you if its infrastructure or application.

    Spin up a Joyent SmartOS or do a local install of OpenIndiana.
    Then install your app

    This d script will show you when garbage collection is starting and how
    long it takes.


    node*:::gc-start
    {
    self->ts = timestamp;
    }

    node*:::gc-done
    {
    printf("Garbage Collection - [%Y] - %d",
    walltimestamp,
    timestamp - self->ts);
    self->ts = 0
    }


    Anton
    On Wednesday, November 14, 2012 9:08:04 AM UTC, Dor Tzur wrote:

    We're having a problem where every once in a while one of our environments
    our node app runs on 100% CPU. The server isn't very active and usually
    runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Ben Noordhuis at Nov 15, 2012 at 2:42 pm ⇧

    On Thu, Nov 15, 2012 at 10:41 AM, Anton Whalley wrote:
    Can you put it on an illumOS based system temporarily and get DTracing?
    At the very least this would tell you if its infrastructure or application.

    Spin up a Joyent SmartOS or do a local install of OpenIndiana.
    Then install your app

    This d script will show you when garbage collection is starting and how long
    it takes.


    node*:::gc-start
    {
    self->ts = timestamp;
    }

    node*:::gc-done
    {
    printf("Garbage Collection - [%Y] - %d",
    walltimestamp,
    timestamp - self->ts);
    self->ts = 0
    }
    In that vein, if you're on a semi-recent RHEL-based system (CentOS,
    Fedora), you can use this systemtap script. It prints aggregated GC
    statistics every 5 seconds and at program exit. Run with `sudo stap
    gc.stp -c 'node script.js'`.

    Note that systemtap support currently only exists in master.
    Debian/Ubuntu probably won't work, the systemtap in the official
    repositories is too old.

    #!/usr/bin/env stap

    global samples
    global all_samples
    global timestamp

    probe process("node").mark("gc__start")
    {
    timestamp = gettimeofday_us()
    }

    probe process("node").mark("gc__done")
    {
    sample = gettimeofday_us() - timestamp
    samples <<< sample
    all_samples <<< sample
    }

    probe timer.s(5)
    {
    print(@hist_log(samples))
    printf("min:%d avg:%d max:%d count:%d\n",
    @min(samples), @avg(samples),
    @max(samples), @count(samples))
    delete samples
    }

    probe end
    {
    print(@hist_log(all_samples))
    printf("min:%d avg:%d max:%d count:%d\n",
    @min(all_samples), @avg(all_samples),
    @max(all_samples), @count(all_samples))
    }

    As a gist: https://gist.github.com/4078925

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Darcy at Nov 16, 2012 at 2:33 am ⇧
    node-webkit-agent is a good tool.
    if you just want to know whether it's caused by gc, the easiest way is run
    your program by "node --trace_gc myapp.js", and you can see GC time like
    this:

    785292 ms: Mark-sweep 9.5 (46.0) -> 9.4 (46.0) MB, 14 ms ...

    where '14ms' means busy doing GC with CPU usage of nearly 100%.
    On Wednesday, November 14, 2012 5:08:04 PM UTC+8, Dor Tzur wrote:

    We're having a problem where every once in a while one of our environments
    our node app runs on 100% CPU. The server isn't very active and usually
    runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
  • Murvin Lai at Nov 16, 2012 at 10:51 pm ⇧
    For your server, how many RAM u have? it may be related to memory.
    try either increase your physical memory, or run node with extra memory
    allocated. (default is 1GB)



    On Thu, Nov 15, 2012 at 6:33 PM, darcy wrote:

    node-webkit-agent is a good tool.
    if you just want to know whether it's caused by gc, the easiest way is run
    your program by "node --trace_gc myapp.js", and you can see GC time like
    this:

    785292 ms: Mark-sweep 9.5 (46.0) -> 9.4 (46.0) MB, 14 ms ...

    where '14ms' means busy doing GC with CPU usage of nearly 100%.
    On Wednesday, November 14, 2012 5:08:04 PM UTC+8, Dor Tzur wrote:

    We're having a problem where every once in a while one of our
    environments our node app runs on 100% CPU. The server isn't very active
    and usually runs on 0%-2% CPU.
    I was wondering what are the common issues that might cause this problem
    and what would be the best way to find out what causing this issue.

    technical spec:
    node version 0.8.14
    ubuntu 11.10
    Intel(R) Xeon(R) CPU E5645 @ 2.40GHz

    gems used:
    "express" : 2.5.x, "log" : "1.2.x", "redis" : "0.8.x", "socket.io" :
    "0.9.x", "mongodb": ">= 0.9.6-7", "passport" : "0.x.x", "passport-local" :
    "0.x.x",

    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines:
    https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en
    --
    Job Board: http://jobs.nodejs.org/
    Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
    You received this message because you are subscribed to the Google
    Groups "nodejs" group.
    To post to this group, send email to [email protected]
    To unsubscribe from this group, send email to
    nodejs+[email protected]
    For more options, visit this group at
    http://groups.google.com/group/nodejs?hl=en?hl=en

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupnodejs @
categoriesnodejs
postedNov 14, '12 at 3:44p
activeNov 16, '12 at 10:51p
posts12
users10
websitenodejs.org
irc#node.js

10 users in discussion

Anand George: 2 posts Dor Tzur: 2 posts Murvin Lai: 1 post Ben Noordhuis: 1 post Jimb Esser: 1 post Darcy: 1 post Anton Whalley: 1 post Adam Crabtree: 1 post Michael Schoonmaker: 1 post Chad Engler: 1 post

Content

People

Support

Translate

site design / logo © 2023 Grokbase