Do you have performance issues on the VPS? [message #38238] |
Tue, 01 December 2009 11:45 |
fernandomm
Messages: 11 Registered: November 2009
|
Junior Member |
|
|
Hello,
Last weekend i tried to move a high traffic website to a VPS on the same hardware that i was running it before.
I have set the ubc and other parameters like this:
# UBC parameters (in form of barrier:limit)
KMEMSIZE="999999999:999999999"
LOCKEDPAGES="256:256"
PRIVVMPAGES="524288:643072"
SHMPAGES="21504:21604"
NUMPROC="2000:2000"
PHYSPAGES="0:9223372036854775807"
VMGUARPAGES="524288:524288"
OOMGUARPAGES="524288:524288"
NUMTCPSOCK="999999999:999999999"
NUMFLOCK="999999999:999999999"
NUMPTY="16:16"
NUMSIGINFO="1024:1024"
TCPSNDBUF="999999999:999999999"
TCPRCVBUF="999999999:999999999"
OTHERSOCKBUF="999999999:999999999"
DGRAMRCVBUF="262144:262144"
NUMOTHERSOCK="999999999:999999999"
DCACHESIZE="3409920:3624960"
NUMFILE="999999999:999999999"
AVNUMPROC="180:180"
NUMIPTENT="800:800"
# Disk quota parameters (in form of softlimit:hardlimit)
DISKSPACE="211048576:211153024"
DISKINODES="400000:440000"
QUOTATIME="0"
# CPU fair sheduler parameter
CPUUNITS="153424"
After a while running it ( it's the same OS, same apache version and conf, same mysql conf etc. ) the apache processes stops working and use a lot of CPU ( load goes up to 400-500 ).
If i strace one of the processes, they do not show anything. Also, the apache logs are not updated the this starts to happen.
It is important to note that user_beancounters do not show any failcount, but the system is unusable.
We have about 400 requests per second, but our current hardware was running fine until me moved to a VPS.
Do you see any configuration issues at my ubc parameters? Could the diskinodes limitation be causing this problem ( limiting apache access to disk and making it goes crazy? )?
Anyway, any opinions and experiences about this issue are welcome
Thanks!
[Updated on: Tue, 01 December 2009 11:45] Report message to a moderator
|
|
|
|
Re: Do you have performance issues on the VPS? [message #38240 is a reply to message #38239] |
Tue, 01 December 2009 12:51 |
fernandomm
Messages: 11 Registered: November 2009
|
Junior Member |
|
|
I get this:
# df -i
Filesystem Inodes IUsed IFree IUse% Mounted on
simfs 5518274 42517 5475757 1% /
This is really weird. If i send 30 concurrent connections my VPS is locked up while the HN is handling more than 500 requests/second current.
ab -n 20000 -c 30 http://myurl.com
Top:
top - 04:48:25 up 1:52, 0 users, load average: 84.37, 20.80, 6.94
Tasks: 254 total, 207 running, 47 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.9%us, 0.4%sy, 0.0%ni, 98.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 2049136k total, 594000k used, 1455136k free, 0k buffers
Swap: 0k total, 0k used, 0k free, 0k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
545 apache 20 0 248m 9796 4744 R 3.9 0.5 0:01.74 httpd
555 apache 20 0 248m 9832 4676 R 3.9 0.5 0:02.42 httpd
738 apache 20 0 248m 9748 4712 R 3.9 0.5 0:01.76 httpd
1352 apache 20 0 248m 9684 4700 R 3.9 0.5 0:01.44 httpd
3844 apache 20 0 248m 9080 4332 R 3.9 0.4 0:00.42 httpd
3847 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.40 httpd
3850 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.38 httpd
3851 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.42 httpd
3853 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.38 httpd
3862 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.38 httpd
3863 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.36 httpd
3871 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.34 httpd
3887 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.36 httpd
3889 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.36 httpd
3890 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.38 httpd
3891 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.34 httpd
3896 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.32 httpd
3897 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.32 httpd
3905 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.32 httpd
3911 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.30 httpd
3915 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.30 httpd
3928 apache 20 0 248m 8944 4248 R 3.9 0.4 0:00.34 httpd
3933 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.32 httpd
3939 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.28 httpd
3940 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.30 httpd
3942 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.30 httpd
3945 apache 20 0 248m 8444 3784 R 3.9 0.4 0:00.28 httpd
3951 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3953 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3955 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3957 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.28 httpd
3961 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.28 httpd
3967 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3968 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3981 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.26 httpd
3983 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.24 httpd
3984 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.24 httpd
3985 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.24 httpd
3999 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.22 httpd
4005 apache 20 0 248m 5564 1428 R 3.9 0.3 0:00.22 httpd
|
|
|
|
|
Re: Do you have performance issues on the VPS? [message #38247 is a reply to message #38245] |
Tue, 01 December 2009 17:19 |
fernandomm
Messages: 11 Registered: November 2009
|
Junior Member |
|
|
Here it is, under a high load situation:
localhost:/# cat /proc/bc/1052/resources
kmemsize 52574232 79876595 999999999 999999999 0
lockedpages 0 0 256 256 0
privvmpages 103875 177876 524288 643072 0
shmpages 4134 4790 21504 21604 0
numproc 129 233 2000 2000 0
physpages 24032 74638 0 9223372036854775807 0
vmguarpages 0 0 524288 524288 0
oomguarpages 32664 87515 524288 524288 0
numtcpsock 70 115 999999999 999999999 0
numflock 7 12 999999999 999999999 0
numpty 0 2 16 16 0
numsiginfo 0 116 1024 1024 0
tcpsndbuf 1220800 2027664 999999999 999999999 0
tcprcvbuf 1146880 215304 999999999 999999999 0
othersockbuf 43040 184088 999999999 999999999 0
dgramrcvbuf 0 4360 262144 262144 0
numothersock 71 197 999999999 999999999 0
dcachesize 492036 689226 3409920 3624960 0
numfile 2469 4210 999999999 999999999 0
numiptent 124 124 800 800 0
localhost:/# cat /proc/bc/1052/ioacct
read 395423744
write 3526656
dirty 3547136
cancel 61440
missed 0
syncs_total 0
fsyncs_total 81
fdatasyncs_total 4
range_syncs_total 0
syncs_active 0
fsyncs_active 0
fdatasyncs_active 0
range_syncs_active 0
vfs_reads 281220
vfs_read_chars 409438689
vfs_writes 89204
vfs_write_chars 16777536
io_pbs 5
localhost:/# cat /proc/bc/1052/vmaux
unused_privvmpages 76180
tmpfs_respages 471
swap_pages 8632
swapin 383847
unmap 87745
[Updated on: Tue, 01 December 2009 18:54] Report message to a moderator
|
|
|