Hi -
I'm trying to configure LVEs optimally. I'm not being very successful. Sites either run out of resources or use too many resources. I'm hoping someone will be willing to share their optimized config. It would sure help me to better understand!
Here's a summary that gives some details about the particular server:
CLOUDLINUX 6.8 x86_64 standard WHM 58.0 (build 28)
Total processors: 8
GenuineIntel
Intel(R) Xeon(R) CPU E5606 @ 2.13GHz
Speed 2133.533 MHz
Cache 8192 KB
Memory: 24497984k/26214400k available
My default ve.cfg (I've removed all the custom entries because I wind up with a slow server!):
<?xml version="1.0" encoding="utf-8"?>
<lveconfig>
<defaults>
<cpu limit="50%"/>
<ncpu limit="1"/>
<io limit="1024"/>
<vmem limit="0"/>
<pmem limit="262144"/>
<nproc limit="100"/>
<other maxentryprocs="20"/>
<mem limit="262144"/>
<iops limit="1024"/>
</defaults>
</lveconfig>
There are right around 140 sites on this server. Most use very little CPU and RAM, but I have about 4 sites that are more resource intensive.
Thanks for your insight!
Beth
I'm trying to configure LVEs optimally. I'm not being very successful. Sites either run out of resources or use too many resources. I'm hoping someone will be willing to share their optimized config. It would sure help me to better understand!
Here's a summary that gives some details about the particular server:
CLOUDLINUX 6.8 x86_64 standard WHM 58.0 (build 28)
Total processors: 8
GenuineIntel
Intel(R) Xeon(R) CPU E5606 @ 2.13GHz
Speed 2133.533 MHz
Cache 8192 KB
Memory: 24497984k/26214400k available
My default ve.cfg (I've removed all the custom entries because I wind up with a slow server!):
<?xml version="1.0" encoding="utf-8"?>
<lveconfig>
<defaults>
<cpu limit="50%"/>
<ncpu limit="1"/>
<io limit="1024"/>
<vmem limit="0"/>
<pmem limit="262144"/>
<nproc limit="100"/>
<other maxentryprocs="20"/>
<mem limit="262144"/>
<iops limit="1024"/>
</defaults>
</lveconfig>
There are right around 140 sites on this server. Most use very little CPU and RAM, but I have about 4 sites that are more resource intensive.
Thanks for your insight!
Beth
Last edited by a moderator: