1 second delay in capturing the sar data.
Issue
- Collecting 1 second sar data, with the cron job:
/usr/lib64/sa/sa1 -S INT 1 50400
For most statistics this reports is correct. However Some busy network interfaces running at 1000/full, and the sar data shows all zero's, every other second...
Example:
$ sar -n DEV | grep eth7
06:00:19 AM eth7 26.00 0.00 1.52 0.00 0.00 0.00 26.00
06:00:20 AM eth7 0.00 0.00 0.00 0.00 0.00 0.00 0.00
06:00:21 AM eth7 26.00 0.00 1.52 0.00 0.00 0.00 26.00
06:00:22 AM eth7 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Presumption is, this is just a "feature" of sar and that it cannot "handle" such throughput of data. Is this correct ?
Environment
- Red Hat Enterprise Linux 5.6
- Kernel-2.6.18-238.9.1.el5
- sysstat-9.0.4-1.x86_64
- igb Drivers
- e1000e Drivers
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.