Unusual high CPU cycles on iRule
Hi, So I am using the following iRule on all virtual servers & just enabled timing on this: priority 50 when RULE_INIT { set static::maindatalist set static::debug 0 } when HTTP_REQUEST { return the company logo from ifile if { ([HTTP::uri] eq "/bip-company-logo.gif") }{ HTTP::respond 200 content [ifile get /Common/company-logo] return } set vs [virtual name] if {($static::debug == 1)}{ log local0. "VS is $vs, looking at $static::maindatalist" } set status [class match -value "[virtual name]" equals "maintenance-list"] if {($static::debug == 1)}{ log local0. "Looked up status: $status" } if { ($status eq "1") } { if { ([HTTP::uri] eq "/favicon.ico") }{ HTTP::respond 404 content "" return } set contact "Application Support () or your Regional Helpdesk" if {([virtual name] starts_with "/ITSS/")}{ set contact "your Regional Helpdesk" } if {($static::debug == 1)}{ log local0. "Showing maintenance page" } HTTP::respond 200 content " Down for maintenance - company Maintenance This application is currently undergoing maintenance. It should be available again within the specified time period. For any questions, please contact $contact. " Connection Close return } } when LB_FAILED { set contact "Application Support () or your Regional Helpdesk" if {([virtual name] starts_with "/ITSS/")}{ set contact "your Regional Helpdesk" } HTTP::respond 503 content " Application Unavailable - company Application unavailable This application is currently not available. Please contact $contact. " Connection Close return } (378 requests) RULE_INIT: 5900 min, 12300 avg, 12300 max LB_FAILED: 46000 min, 60200 avg, 167000 max HTTP_REQUEST: 21300 min, 44000 avg, 384900 max I used the F5DevCentral_iRulesRuntimeCalculator to calculate the CPU usage, coming down to 28% max CPU usage per request. This seems VERY high. Is there anything I can do about this? I tested with a healthy node & a disabled node, no maintenance mode.
iRule to log all HTTP_REQUEST and insert a cookie
Hello, I wrote an iRule that uses HSL for sending syslog level data to remote servers. The iRule sends all HTTP_REQUESTs to HSL pool. There are three pools defined - each for specific purposes. I find that the iRule causes an issue of running on one CPU probably because the way the variables are defined and accessed. I have attached the iRule for your reference. Can somebody tell me if I am doing anything inefficient? Much appreciated. when RULE_INIT { array set ::active_clients { } } when CLIENT_ACCEPTED { set hsl [HSL::open -proto UDP -pool syslog_server_pool] set hsl1 [HSL::open -proto UDP -pool syslog_server_pool_1] set hsl2 [HSL::open -proto UDP -pool syslog_server_pool_2] Send the data using local7 facility <190> HSL::send $hsl "<190> [IP::client_addr] [[clock format [clock seconds] -format "%d/%b/%Y:%H:%M:%S %z"]] --[IP::local_addr]--NY--80--CLIENT_ACCEPTED" } when HTTP_REQUEST { if { [info exists ::active_clients([IP::client_addr])] } { incr ::active_clients([IP::client_addr]) } else { set ::active_clients([IP::client_addr]) 1 } if { not ([HTTP::cookie exists SITESERVER] ) } { set F2C "[string range [AES::key 128] 15 end][string range [AES::key 128] 15 end]" HTTP::cookie insert name "F2C" value $F2C } HSL::send $hsl1 "<190> [IP::client_addr] \[[clock format [clock seconds] -format "%d/%b/%Y:%H:%M:%S %z"]\] --[IP::local_addr]--NY--80--[URI::path [HTTP::uri]][URI::basename [HTTP::uri]]--UA=[HTTP::header User-Agent]--CT=[HTTP::header "Content-type"]" HSL::send $hsl2 "<190> [IP::client_addr] \[[clock format [clock seconds] -format "%d/%b/%Y:%H:%M:%S %z"]\] --[IP::local_addr]--NY--80--P2--[URI::path [HTTP::uri]][URI::basename [HTTP::uri]]--QS=[HTTP::query]--C=[HTTP::header cookie]--UA=[HTTP::header User-Agent]--CT=[HTTP::header "Content-type"]" } when HTTP_REQUEST_DATA { set refid "unknown" foreach x [split [string tolower [HTTP::payload]] "&"] { if { $x starts_with "refid=" } { set refid [lindex [split $x "="] 1] } } HSL::send $hsl2 "<190> [IP::client_addr] [[clock format [clock seconds] -format "%d/%b/%Y:%H:%M:%S %z"]] --[IP::local_addr]--NY--80--P2--refid=$refid--F2C=$F2C" } when HTTP_RESPONSE { if { [info exists F2C] and $F2C ne ""} { HTTP::cookie insert name "F2C" value $F2C unset -nocomplain F2C } } when CLIENT_CLOSED { if { [info exists ::active_clients([IP::client_addr])] } { HSL::send $hsl "<190> [IP::client_addr] [[clock format [clock seconds] -format "%d/%b/%Y:%H:%M:%S %z"]] --[IP::local_addr]--NY--80--CLIENT_CLOSED--$::active_clients([IP::client_addr])" unset ::active_clients([IP::client_addr]) } }
High CPU utilisation for process 'kwolker' of even cores
Top output top - 10:47:36 up 16 days, 14:56,1 user,load average: 6.61, 6.60, 6.55 Tasks: 348 total,1 running, 344 sleeping,0 stopped,3 zombie %Cpu0: 98.3 us,1.7 sy,0.0 ni,0.0 id,0.0 wa,0.0 hi,0.0 si,0.0 st %Cpu1:4.6 us,2.6 sy,0.0 ni, 78.4 id,0.3 wa,0.0 hi, 14.1 si,0.0 st %Cpu2: 98.3 us,1.7 sy,0.0 ni,0.0 id,0.0 wa,0.0 hi,0.0 si,0.0 st %Cpu3:2.7 us,2.7 sy,0.0 ni, 94.6 id,0.0 wa,0.0 hi,0.0 si,0.0 st KiB Mem : 21515168 total,136644 free, 19177572 used,2200952 buff/cache KiB Swap:1048572 total,636212 free,412360 used.279056 avail Mem PID USERPRNIVIRTRESSHR S%CPU %MEMTIME+ COMMAND 12795 root10 -10 29651722.0g1112 S63.59.910305:47 kwolker 29071 root10 -10 29651722.0g1184 S63.19.916356:45 kwolker 20101 root10 -10827056 2641761308 S62.81.28902:54 kwolker 21192 root1 -1913.0g1.6g1.6g S10.37.9 795:25.12 tmm.0
Virtual BIG-IQ uses a lot of cpu
Hi Is it normal that a virtual BIG-IQ uses almost 7ghz from a esxi host. Also the cpu is always at around 80-85 % if I check the cpu under monitor on the server itself. Its only used to manage licenses and not BIG-IP systems so it should idle most of the time. Best regards Daniel
High CPU utilization : tmm.0 118.0%
Hi All , Is there any option to find which process is using the more cpu , as we can see for tmm.0 it shows 118% , can we find out what causing the tmm.0 to go up ? Swap: 5242876k total, 0k used, 5242876k free, 2956952k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 18298 root 1 -19 8549m 82m 61m R 118.8 0.5 79279:13 tmm.0 21723 root 39 19 195m 128m 8720 S 74.3 0.8 1:30.87 java 19329 root 20 0 2342m 277m 24m S 53.9 1.7 3389:31 bd 13582 root 20 0 48936 29m 15m S 36.5 0.2 67574:52 bigd 19346 root 20 0 115m 46m 31m S 29.7 0.3 421:18.84 pabnagd 6989 root 20 0 177m 117m 25m S 3.6 0.7 6823:17 mcpd 17797 root 20 0 175m 151m 20m S 3.6 0.9 4175:09 asm_start 5672 root 25 5 113m 27m 22m S 1.6 0.2 3715:39 merged cpu_id five_sec_avg.ratio ------ ------------------ 0 72 1 96 2 71 3 84 -------------------------- Sys::TMM: 0.0 -------------------------- Global TMM Process Id 18298 Running TMM Id 0 TMM Count 1 CPU Id 0 Memory (bytes) Total 6.5G Used 3.3G CPU Usage Ratio (%) Last 5 Seconds 61 Last 1 Minute 59 Last 5 Minutes 57 -------------------------- Sys::TMM: 0.2 -------------------------- Global TMM Process Id 18298 Running TMM Id 2 TMM Count 1 CPU Id 2 Memory (bytes) Total 0 Used 0 CPU Usage Ratio (%) Last 5 Seconds 53 Last 1 Minute 51 Last 5 Minutes 50
Which CPU is the Analysis plane on F5 chassis ?
Based on this article https://support.f5.com/csp/article/K15468, All even CPUs are TMM, All odd CPUs are other process except the last one(analysis plane). Is the article valid for F5 chassis. I would like to know which blade's CPU on a chassis handle analysis plane ? Example My F5 CPUs F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".1 = Gauge32: 6 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".2 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".3 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".4 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".5 = Gauge32: 11 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".6 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".7 = Gauge32: 15 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".8 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".9 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".10 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".11 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".12 = Gauge32: 5 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".13 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".14 = Gauge32: 2 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".15 = Gauge32: 14 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".16 = Gauge32: 4 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".17 = Gauge32: 13 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".18 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".19 = Gauge32: 13 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".20 = Gauge32: 0 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".21 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".22 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".23 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."1".24 = Gauge32: 0 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".1 = Gauge32: 4 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".2 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".3 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".4 = Gauge32: 7 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".5 = Gauge32: 14 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".6 = Gauge32: 2 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".7 = Gauge32: 11 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".8 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".9 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".10 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".11 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".12 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".13 = Gauge32: 15 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".14 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".15 = Gauge32: 14 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".16 = Gauge32: 2 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".17 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".18 = Gauge32: 3 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".19 = Gauge32: 11 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".20 = Gauge32: 8 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".21 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".22 = Gauge32: 1 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".23 = Gauge32: 10 F5-BIGIP-SYSTEM-MIB::sysMultiHostCpuUser5s."3".24 = Gauge32: 0 Thanks,Unexpected CPU utilization on LTM
Hello I am testing 2 different LTMs(VE) and would like to compere the CPU performance to select CPUs for our customer enviroment. BIG-IP 12.1.1 Build 0.0.184 Final 8core BIG-IP 13.1.1 Build 0.0.4 Final 12core During that load testing(1000TPS for SSL), I noticed that the CPU usage of Lower version LTM(8cores) is at 40%, but higher version LTM(12cores) is at 50%. I expected that 12 cores LTM had lower consumption of CPU than 8cores, these results were completely the opposit. Could someone tell me why did that turn out like this ? Thanks in advance for the help.
Relation between CPU , core and TMM
I have a doubt between CPU , core and TMM. I read that CMP feature allows platforms that contain multiple CPUs to run a single TMM process for every CPU, and multiple TMM threads per processing core. So i have a VIPRION 2250 blade which has : Cores per slot is 20, running only 1 VCMP guest and only 1 slot. According to https://support.f5.com/csp/article/K14358 , we have the below: 1 Deca-core (+HT) - Processor 1 - TMM processer 10 - TMM threads (11.5.x) tmsh output shows 10 separate TMM instances (threads) running Questions: 1. In the device above how does TMM process, TMM threads, processor and CPU collaborate? 2. I also see in the performance tab in CPU Utilization , I get 4 graphs with 4 core each, how is this calculated? Thanks!