<html dir="ltr">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=windows-1256">
<style id="owaParaStyle" type="text/css">P {margin-top:0;margin-bottom:0;}</style>
</head>
<body ocsi="0" fpstyle="1">
<div style="direction: ltr;font-family: Courier New;color: #000000;font-size: 10pt;">
Hello,<br>
<br>
I don't know if this behavior/bug was already reported on this ML, so in doubt.<br>
<br>
Context:<br>
<br>
- SpectrumScale 4.2.2-3<br>
- client node with 64 cores<br>
- OS: RHEL7.3<br>
<br>
When a MPI job with 64 processes is launched on the node with 64 cores then the FS freezed (only the output log file of the MPI job is put on the GPFS; so it may be related to the 64 processes writing in a same file???).<br>
<br>
strace -p 3105 # mmfsd pid stucked<br>
Process 3105 attached<br>
wait4(-1, # stucked at this point<br>
<br>
strace ls /gpfs<br>
stat("/gpfs", {st_mode=S_IFDIR|0755, st_size=131072, ...}) = 0<br>
openat(AT_FDCWD, "/gpfs", O_RDONLY|O_NONBLOCK|O_DIRECTORY|O_CLOEXEC # stucked at this point<br>
<br>
I have no problem with the other nodes of 28 cores.<br>
The GPFS command mmgetstate is working and I am able to use mmshutdown to recover the node.<br>
<br>
<br>
If I put workerThreads=72 on the 64 core node then I am not able to reproduce the freeze and I get the right behavior.<br>
<br>
Is this a known bug with a number of cores > workerThreads?<br>
<br>
<div>
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px">
<div style="font-family:Tahoma; font-size:13px"><font face="Courier New">Best regards,<br>
<img alt="-- "><br>
<font style="font-size:9pt" size="2"><b>Nicolas Capit</b></font></font><br>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>