cc1plus consuming CPU

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
roblew
Charter Member
Charter Member
Posts: 123
Joined: Mon Mar 27, 2006 7:32 pm
Location: San Ramon

cc1plus consuming CPU

Post by roblew »

Hello all,

I did a quick search, and was surprised to not find any threads on this topic. I've recently noticed that our RedHat Linux dataStage server is pegged quite often with 0% idle CPU. During this time, I noticed that often, there are one or more "cc1plus" processes comsuming a lot of CPU.

top output:

Code: Select all

 18:59:34  up 19 days,  7:16, 10 users,  load average: 7.59, 7.44, 6.63
433 processes: 423 sleeping, 10 running, 0 zombie, 0 stopped
CPU states:  cpu    user    nice  system    irq  softirq  iowait    idle
           total   83.0%    0.0%   15.0%   0.4%     1.4%    0.0%    0.0%
           cpu00   77.3%    0.0%   18.8%   0.9%     2.8%    0.0%    0.0%
           cpu01   88.6%    0.0%   11.3%   0.0%     0.0%    0.0%    0.0%
Mem:  8465792k av, 5559812k used, 2905980k free,       0k shrd,  368396k buff
                   3060140k actv, 1726140k in_d,    5388k in_c
Swap: 6291352k av,    3564k used, 6287788k free                 3815356k cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME CPU COMMAND
 2923 clph      25   0  285M 285M  3652 R    35.2  3.4  21:57   0 cc1plus
 6011 clph      25   0  193M 193M  3476 R    27.2  2.3   7:42   1 cc1plus
20929 cyau      22   0  7232 7232  5692 R    23.5  0.0   0:07   0 dsapi_slave
11061 cgey      20   0 15432  15M 11304 S     5.1  0.1   0:00   1 osh
We have a large number of developers (~80 concurrent users). I'm wondering if this attributed to running jobs or compiling DataStage jobs or something else? We're running RHEL AS3, with DS 7.5.1A on a 2cpu DL380 conductor node (8GB RAM), clustered with another DL380 and a DL385.

Any ideas?
roblew
Charter Member
Charter Member
Posts: 123
Joined: Mon Mar 27, 2006 7:32 pm
Location: San Ramon

Post by roblew »

I have confirmed that the user "clph" has been compiling two jobs, which have taken longer than usual to compile. One of the job's compilation has been running for around 1 hour thus far.
DSguru2B
Charter Member
Charter Member
Posts: 6854
Joined: Wed Feb 09, 2005 3:44 pm
Location: Houston, TX

Post by DSguru2B »

Thats one heck of a job your developer has there :shock:
Creativity is allowing yourself to make mistakes. Art is knowing which ones to keep.
roblew
Charter Member
Charter Member
Posts: 123
Joined: Mon Mar 27, 2006 7:32 pm
Location: San Ramon

Post by roblew »

found another compiler process...

Code: Select all

 20:54:52  up 19 days,  9:11, 12 users,  load average: 9.18, 5.95, 3.40
400 processes: 394 sleeping, 6 running, 0 zombie, 0 stopped
CPU states:  cpu    user    nice  system    irq  softirq  iowait    idle
           total   87.2%    0.0%    6.4%   0.0%     1.9%    0.2%    3.9%
           cpu00   83.3%    0.0%    7.7%   0.0%     3.9%    0.3%    4.5%
           cpu01   91.2%    0.0%    5.1%   0.0%     0.0%    0.1%    3.3%
Mem:  8465792k av, 4429968k used, 4035824k free,       0k shrd,  377036k buff
                   2700936k actv,  910984k in_d,   49764k in_c
Swap: 6291352k av,    2924k used, 6288428k free                 3052388k cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME CPU COMMAND
14942 scvd      25   0 98.9M  98M  3328 R    35.9  1.1   3:37   1 cc1plus
15141 hiqu      25   0 16996  16M 11916 S    28.5  0.2   1:47   1 osh
15139 hiqu      16   0 15580  15M 11684 S    12.7  0.1   0:36   0 osh
15145 hiqu      25   0 17756  17M  4112 R     5.6  0.2   0:18   1 sqlldr
not sure what's going on here... not sure if the gcc version is relevent here: gcc version 3.2.3 20030502 (Red Hat Linux 3.2.3-52)
Post Reply