Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

High CPU utilization on idle mount. #62

Closed
bedge opened this issue Mar 3, 2016 · 4 comments
Closed

High CPU utilization on idle mount. #62

bedge opened this issue Mar 3, 2016 · 4 comments

Comments

@bedge
Copy link
Contributor

bedge commented Mar 3, 2016

I'm seeing 2 cores maxed out by the goofys daemon.

Mount opts are:

123.bucketname.xxx.com on /content-goofys type fuse (rw,nosuid,nodev,default_permissions,allow_other)

top shows 2 threads maxed out:

top - 17:13:33 up 1 day, 18:38,  8 users,  load average: 4.08, 4.33, 5.07
Tasks: 190 total,   1 running, 188 sleeping,   0 stopped,   1 zombie
%Cpu(s): 50.5 us,  0.2 sy,  0.0 ni, 49.3 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem:  16433132 total, 13873464 used,  2559668 free,   245852 buffers
KiB Swap:        0 total,        0 used,        0 free.  8638932 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
10801 root      20   0 7016036 2.715g   4276 S 200.1 17.3   1556:48 goofys

version is:

goofys version 0.0.4

terminating & restarting the mount resolved the CPU issue. Not sure what caused it in the 1st place.

@kahing
Copy link
Owner

kahing commented Mar 3, 2016

are there other processes reading from the mountpoint at the time? Also please upgrade to 0.0.5 as that fixes a deadlock issue (probably not related to your problem though)

@kahing
Copy link
Owner

kahing commented Mar 3, 2016

and if you could attach a debugger and see what it was doing, that would be even better

@bedge
Copy link
Contributor Author

bedge commented Mar 3, 2016

At the time the connection was idle. All goofys clients had timed out and terminated
I have upgraded to 0.0.5 since this was reported.
Will be re-running tests with the newer version.

One thing I can say is that there are many concurrent threads accessing the mount point when the tests are running. So if this hasn't been tested with a high level of concurrency, that may be related.

If it happens again, it's in AWS, so I could get you access to the VM if you want to see first hand.

@bedge
Copy link
Contributor Author

bedge commented Mar 7, 2016

Have not been able to replicate with 0.0.5 running high volume.

@bedge bedge closed this as completed Mar 7, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants