Does the box have hypertheading enabled? I've seen some wired behavior ocassionaly with hypertheading because it essentially makes the output of top a little deceptive.
The other thing is what category it the cpu utilization for instance it's not uncommon for a core to go to 100% due to io wait (wa in top). In this case a change of the scheduler may help. I often find the default which is cfq is not right for servers. Sent from my BlackBerry 10 smartphone. Original Message From: Ian A Taylor Sent: Thursday, January 22, 2015 12:23 To: Stijn De Weirdt; Scientific Linux Users Subject: Re: cpu process load jumps to 100% Stijn Thanks for reply The last 3 times it has been events/17 That has jumped to 100 utilization Is this saying processor 17 is flaky ? How would I test that ? Regards Ian t On 22/01/15 17:13, Stijn De Weirdt wrote: > hi ian, > > run top to see which process is causing the load. > > if it is kipmi, try to add this to the active kernel entry in > /etc/grub.conf > ipmi_si.kipmid_max_busy_us=100 > and reboot if you can. if you can't reboot, there's probably a way to > write to /sys to set it (but it will be non-permanent). > > stijn > > > > On 01/22/2015 05:48 PM, Ian A Taylor wrote: >> Sir/Madam >> >> I have a 64bit machine on which I run SL6 >> >> I upgraded to the latest kernel >> >> uname -a >> Linux AAAAAAAA 2.6.32-504.3.3.el6.x86_64 #1 SMP Tue Dec 16 14:29:22 CST >> 2014 x86_64 x86_64 x86_64 GNU/Linux >> >> Since I have done this then the machine behaves ok for 25 mins then >> >> The process load on one of the CPU jumps up to 100% >> >> There are 40 processor on this machine >> >> >> processor : 39 >> vendor_id : GenuineIntel >> cpu family : 6 >> model : 62 >> model name : Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz >> stepping : 4 >> microcode : 1046 >> cpu MHz : 2499.983 >> cache size : 25600 KB >> physical id : 1 >> siblings : 20 >> core id : 12 >> cpu cores : 10 >> apicid : 57 >> initial apicid : 57 >> fpu : yes >> fpu_exception : yes >> cpuid level : 13 >> wp : yes >> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca >> cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall >> nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good >> xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx >> smx est tm2 ssse3 cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic popcnt >> tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm ida arat epb >> xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid fsgsbase smep >> erms >> bogomips : 4999.35 >> clflush size : 64 >> cache_alignment : 64 >> address sizes : 46 bits physical, 48 bits virtual >> power management: >> >> >> >> >> There is nothing to indicate why this is happening in /var/log/messages >> >> Originally I had a problem with kondemand but I solved this by turning >> off service cpuspeed >> >> However I am not clear why the process load jumps up to 100% >> >> Any help would be sincerely appreciated >> >> >> >> Thanking you. >> >> Yours sincerely >> >> >> >> Ian Taylor >> University of St.Andrews, >> School of Physics & Astronomy, >> North Haugh, >> St.Andrews, >> Fife KY16 9SS, >> Scotland. >> >> e-Mail :- [email protected] >> Tel :- (0)1334-463141 >> Fax :- (0)1334-463104 >> >> The University of St Andrews >> is a charity registered in >> Scotland : No SC013532. >> >> > > -- Thanking you. Yours sincerely Ian Taylor University of St.Andrews, School of Physics & Astronomy, North Haugh, St.Andrews, Fife KY16 9SS, Scotland. e-Mail :- [email protected] Tel :- (0)1334-463141 Fax :- (0)1334-463104 The University of St Andrews is a charity registered in Scotland : No SC013532.
