os.stat bug?

[Laszlo Nagy]

Hi All,

I have a Python program that goes up to 100% CPU. Just like this (top):

PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU
COMMAND
80212 user1 2 44 0 70520K 16212K select 1 0:30 100.00%
/usr/local/bin/python process_updates_ss_od.py -l 10

I have added extra logs and it turns out that there are two threads. One
thread is calling "time.sleep()" and the other is calling "os.stat"
call. (Actually it is calling os.path.isfile, but I hunted down the last
link in the chain.) The most interesting thing is that the process is in
"SELECT" state. As far as I know, CPU load should be 0% because "select"
state should block program execution until the I/O completes.

I must also tell you that the os.stat call is taking long because this
system has about 7 million files on a slow disk. It would be normal for
an os.stat call to return after 10 seconds. I have no problem with that.
But I think that the 100% CPU is not acceptable. I guess that the code
is running in kernel mode. I think this because I can send a KILL signal
to it and the state changes to the following:


PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU
COMMAND
80212 user1 2 44 0 70520K 15256K STOP 5 1:27 100.00%
/usr/local/bin/python process_updates_ss_od.py -l 10

So the state of the process changes to "STOP", but the program does not
stop until the os.stat call returns back (sometimes for 30 seconds).

Could it be a problem with the operation system? Is it possible that an
os.stat call requires 100% CPU power from the OS? Or is it a problem
with the Python implementation?

(Unfortunately I cannot give you an example program. Giving an example
would require giving you a slow I/O device with millions of files on it.)

OS version: FreeBSD 8.1-STABLE amd64
Python version: 2.6.6

Thanks,

Laszlo

 

[Martin v. Loewis]

Could it be a problem with the operation system? Is it possible that an

os.stat call requires 100% CPU power from the OS? Or is it a problem
with the Python implementation?

There is a chance that the CPU usage actually comes from the thread
doing sleep(). If you have a very short sleep time, and a loop around
it, it may spin in this loop.

If it's not that, and if it's not any other unrelated application that
uses CPU that you didn't mention, then chances are high that it's indeed
the file system code of your operating system that consumes that much
CPU time.

Regards,
Martin

 

[Laszlo Nagy]

There is a chance that the CPU usage actually comes from the thread
doing sleep(). If you have a very short sleep time, and a loop around
it, it may spin in this loop. Sleep time is 1.0 sec.
If it's not that, and if it's not any other unrelated application that
uses CPU that you didn't mention, then chances are high that it's indeed
the file system code of your operating system that consumes that much
CPU time.

Yes. top shows 2 thread in this process, and full 100% CPU time is
dedicated to the process. So it must be it.

Thank you,

Laszlo

 

[Nobody]

So the state of the process changes to "STOP", but the program does not
stop until the os.stat call returns back (sometimes for 30 seconds).

Could it be a problem with the operation system? Is it possible that an
os.stat call requires 100% CPU power from the OS? Or is it a problem
with the Python implementation?

It's the OS kernel. If it was Python or the C library, sending SIGKILL
would result in immediate termination.

Is the disk interface operating in PIO mode? A slow disk shouldn't cause
100% CPU consumption; the OS would just get on with something else (or
just idle) while waiting for data to become available. But if it's
having to copy data from the controller one word at a time, that could
cause it (and would also make the disk appear slow).

 

[Laszlo Nagy]

It's the OS kernel. If it was Python or the C library, sending SIGKILL
would result in immediate termination.

Is the disk interface operating in PIO mode? A slow disk shouldn't cause
100% CPU consumption; the OS would just get on with something else (or
just idle) while waiting for data to become available. But if it's
having to copy data from the controller one word at a time, that could
cause it (and would also make the disk appear slow).

This is a RAID 1+0 array with 10 hard disks plus a SCSI controller with
2GB write back cache. I don't think that it has to do anything with disk
speed. The CPU load goes up to 100% (the disk I/O does not go up that much).

I'm working about a different storage method. We will be storing these
logs in a real database instead of separate CSV files. So probably this
problem will cease.

Thanks,

Laszlo