Solaris Monitoring: различия между версиями
Материал из noname.com.ua
Перейти к навигацииПерейти к поискуSirmax (обсуждение | вклад) |
Sirmax (обсуждение | вклад) |
||
Строка 71: | Строка 71: | ||
# LAT – waiting for a processor (sitting on a run queue) |
# LAT – waiting for a processor (sitting on a run queue) |
||
− | <PRE>prstat -m</ |
+ | <PRE>prstat -m</PRE> |
Use prstat -m LAT category, in conjunction with utilization measurements, delivered workload throughput and run queue depth (vmstat “r” column) to determine for CPU capacity planning |
Use prstat -m LAT category, in conjunction with utilization measurements, delivered workload throughput and run queue depth (vmstat “r” column) to determine for CPU capacity planning |
||
Строка 101: | Строка 101: | ||
Total: 711 processes, 902 lwps, load averages: 3.84, 4.30, 4.37 |
Total: 711 processes, 902 lwps, load averages: 3.84, 4.30, 4.37 |
||
</PRE> |
</PRE> |
||
− | |||
==Solaris iostat== |
==Solaris iostat== |
Версия 16:21, 1 марта 2011
pargs pflags pcred pldd psig pstack pmap pfiles
pwdx pstop prun
pwait ptree ptime preap pkill
trapstat
mpstat vmstat netstat sar
cputrack cpustat
fsstat
Microstates
- Fine-grained state tracking for processes/threads
- Off by default in Solaris 8 and Solaris 9
- On by default in Solaris 10
- Can be enabled per-process via /proc
- prstat -m reports microstates
- As a percentage of time for the sampling period
- USR – user mode
- SYS - kernel mode
- TRP – trap handling
- TFL – text page faults
- DFL – data page faults
- LCK – user lock wait
- SLP - sleep
- LAT – waiting for a processor (sitting on a run queue)
prstat -m
Use prstat -m LAT category, in conjunction with utilization measurements, delivered workload throughput and run queue depth (vmstat “r” column) to determine for CPU capacity planning
prstat – user summary
# prstat -t NPROC USERNAME SIZE RSS MEMORY TIME CPU 128 root 446M 333M 1.4% 47:14:23 11% 2 measter 6600K 5016K 0.0% 0:00:07 0.2% 1 clamb 9152K 8344K 0.0% 0:02:14 0.1% 2 rmc 7192K 6440K 0.0% 0:00:00 0.1% 1 bricker 5776K 4952K 0.0% 0:00:20 0.1% 2 asd 10M 8696K 0.0% 0:00:01 0.1% 1 fredz 7760K 6944K 0.0% 0:00:05 0.1% 2 jenks 8576K 6904K 0.0% 0:00:01 0.1% 1 muffin 15M 14M 0.1% 0:01:26 0.1% 1 dte 3800K 3016K 0.0% 0:00:04 0.0% 2 adjg 8672K 7040K 0.0% 0:00:03 0.0% 3 msw 14M 10M 0.0% 0:00:00 0.0% 1 welza 4032K 3248K 0.0% 0:00:29 0.0% 2 kimc 7848K 6344K 0.0% 0:00:25 0.0% 4 jcmartin 13M 9904K 0.0% 0:00:03 0.0% 1 rascal 17M 16M 0.1% 0:02:11 0.0% 1 rab 3288K 2632K 0.0% 0:02:11 0.0% 1 gjmurphy 3232K 2392K 0.0% 0:00:00 0.0% 1 ktheisen 15M 14M 0.1% 0:01:16 0.0% 1 nagendra 3232K 2400K 0.0% 0:00:00 0.0% 2 ayong 8320K 6832K 0.0% 0:00:02 0.0% Total: 711 processes, 902 lwps, load averages: 3.84, 4.30, 4.37
Solaris iostat
● Wait: number of threads queued for I/O ● Actv: number of threads performing I/O ● wsvc_t: Average time spend waiting on queue ● asvc_t: Average time performing I/O ● %w: Only useful if one thread is running on the entire machine – time spent waiting for I/O ● %b: Device utilization – only useful if device can do just 1 I/ O at a time (invalid for arrays etc...)
# iostat -xnz extended device statistics r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 687.8 0.0 38015.3 0.0 0.0 1.9 0.0 2.7 0 100 c0d0 Queue Performing I/O wait svc
New Formatting flags -C, -l, -m, -r, -s, -z, -T ● -C: report disk statistics by controller ● -l n: Limit the number of disks to n ● -m: Display mount points (most useful with -p) ● -r: Display data n comma separated format ● -s: Suppress state change messages ● -z: Suppress entries with all zero values ● -T d|u Display a timestamp in date (d) or unix time_t (u)