need help for Memory utilization and Network speed change
Hello ALL,
Environment:
master,media server (Redhat) Netbackup 7.1.
Client (HPUX) Netbackup 7.1
VTL (fujitsu CS800)
Issue:
When backup process starts on client (as per schedul) affecting its performance.
Can you clarify about how to change the buffer size and decrease the network speed ex.20MB/s
I found the article about similar issue on forum here https://www-secure.symantec.com/connect/forums/netbackup-memory-utilization
.But on our client machine under /usr/openv/netbackup/db/config/ is empty.
The following is client machines cpu/ memory info.
bash-3.2# uname -a
HP-UX gcrmdb01 B.11.31 U ia64 2139938236 unlimited-user license
bash-3.2# top
System: gcrmdb01 Tue Feb 4 18:09:02 2014
Load averages: 0.01, 0.00, 0.00
236 processes: 192 sleeping, 44 running
Cpu states:
CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS
0 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
1 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
2 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
3 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
4 0.01 0.0% 0.0% 4.0% 96.0% 0.0% 0.0% 0.0% 0.0%
5 0.01 1.0% 0.0% 0.0% 99.0% 0.0% 0.0% 0.0% 0.0%
6 0.00 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
7 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
--- ---- ----- ----- ----- ----- ----- ----- ----- -----
avg 0.01 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
System Page Size: 4Kbytes
Memory: 4148868K (1624192K) real, 5758612K (2763152K) virtual, 266148K free Page# 1/11
CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND
4 ? 99 root 152 20 2304K 2048K run 0:33 0.63 0.62 pagezerod
2 ? 95 root 152 20 2016K 1792K run 48:16 0.50 0.50 vxfsd
0 ? 20599 oracle 152 20 563M 101M run 0:47 0.43 0.43 java
2 ? 1920 root 152 20 194M 7180K run 2:33 0.31 0.31 nbrmms
2 ? 1585 root 152 20 173M 5024K run 0:49 0.23 0.23 cimprovagt
2 ? 97 root 152 20 648K 576K run 3:05 0.21 0.21 pm_schedcpu
0 ? 8 root 152 20 1008K 896K run 0:07 0.18 0.18 ksyncer_daemon
6 ? 86 root 152 20 2304K 2048K run 0:01 0.16 0.16 fb_daemon
6 ? 1583 cimsrvr 152 20 59172K 12288K run 16:19 0.14 0.14 cimservermain
5 ? 1587 root 152 20 39012K 544K run 9:09 0.14 0.14 cimprovagt
5 ? 1577 root 152 20 25872K 1892K run 0:01 0.14 0.14 rpcd
3 ? 1967 root 152 20 98780K 1436K run 0:11 0.14 0.14 nbsvcmon
2 ? 2171 root 152 20 85516K 4188K run 0:25 0.14 0.14 vxsvc
4 ? 8772 oracle 152 20 109M 14648K run 10:22 0.12 0.12 emagent
2 ? 5730 oracle 154 20 7755M 5572K sleep 0:00 0.12 0.11 ora_j000_crmdb
6 ? 1942 root 152 20 267M 2528K run 0:05 0.10 0.10 nbsl
4 ? 1999 root 152 20 22980K 1460K run 0:04 0.08 0.08 swagentd
3 ? 2296 root 154 20 9896K 496K sleep 0:21 0.08 0.08 kcmond
2 ? 1167 root 152 20 12156K 632K run 0:36 0.07 0.07 automountd
0 ? 16359 oracle 154 20 7758M 3784K sleep 0:18 0.07 0.07 oraclecrmdb
1 ? 20622 oracle 154 20 7755M 2424K sleep 0:07 0.07 0.07 oraclecrmdb
bash-3.2#
bash-3.2# print_manifest |grep -i memory
Main Memory: 16315 MB
bash-3.2#
bash-3.2# swapinfo -tam
Mb Mb Mb PCT START/ Mb
TYPE AVAIL USED FREE USED LIMIT RESERVE PRI NAME
dev 8192 247 7945 3% 0 - 1 /dev/vg00/lvol2
dev 25600 247 25353 1% 0 - 1 /dev/vg00/lvol10
reserve - 8146 -8146
memory 15519 10181 5338 66%
total 49311 18821 30490 38% - 0 -
bash-3.2#
Buffer tuning and network throttle are two different things. Network throttle is describes in these Tech notes:
http://www.symantec.com/docs/HOWTO13915
http://www.symantec.com/docs/TECH153326
http://www.symantec.com/docs/TECH30467
But in general normal backup should not affect system performance - if it does there is something wrong that need to be fixed.
Buffer tuning is used to improve general peformance - it is not a tool to slow down backups.
I wrote a small blog about buffer tuning here: http://www.mass.dk/netbackup/guides/49-netbackup-buffer-tuning.html