Se da un sistem cu 2 procesoare PIII,
memorie destula si cu Red Hat 9 pe el.
# uname -r
2.4.20-18.9smp
# rpm -qa | grep -i postgresql
postgresql-server-7.3.4-1PGDG
postgresql-7.3.4-1PGDG
postgresql-libs-7.3.4-1PGDG
# netstat -n | grep ESTABLISHED | grep ':5432 ' | wc -l
12
load average: 5.89, 4.70, 3.46
Procesele postmaster isi pierd majoritatea timpului
apeland functia sistem "semop" avand doar cateva conexiuni,
dupa cum se vede si mai jos, iar load average "zboara"
in sus. Urmarit cu strace vad apeluri sistem "normale",
apoi apar cateva zeci bune de apeluri "semop" si tot asa.
Vad ca google nu spune mai nimic. S-a mai lovit cineva
de problema asta?
Thanks!
Vitalie
# strace -c -p PID
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
55.15 0.271690 140 1947 semop
20.93 0.103118 1146 90 munmap
8.43 0.041514 1297 32 recv
5.26 0.025926 720 36 mremap
2.73 0.013457 449 30 write
2.18 0.010735 1193 9 read
2.15 0.010594 74 144 brk
1.34 0.006599 103 64 rt_sigaction
0.67 0.003319 184 18 9 stat64
0.57 0.002806 31 90 mmap2
0.37 0.001833 57 32 send
0.12 0.000609 12 52 _llseek
0.04 0.000204 8 27 gettimeofday
0.04 0.000173 10 18 time
0.02 0.000074 8 9 rt_sigprocmask
------ ----------- ----------- --------- --------- ----------------
100.00 0.492651 2598 9 total
---
Detalii despre listele noastre de mail: http://www.lug.ro/
