Professional Documents
Culture Documents
Troubleshooting-Oss Compress
Troubleshooting-Oss Compress
Troubleshooting-Oss Compress
#####################################
It should normally not be necessary to restart the ipcdir_server manually. Pleas
e see the on-line manual page for ipcdir_server.
The Directory Table for CAP_IPC is defined in the ndbm files and is called .ipcd
ir_ndbm_table. To find out the actual directory, enter the command:
# tbs_env get TBS_DAT
If these files are accidentally deleted, the ipcdir_server must be restarted. Ho
wever, in this case all the directory information for the interprocess communica
tion in the system has disappeared; therefore, a complete start-up of the system
should be done.
If the Directory Table has been corrupted, and this cannot be repaired by using
the ipcdir_cleanup process, the Directory Table has to be manually deleted befor
e the system is restarted. The symptom of a corrupted Directory Table is that so
me or all of the TBS managed processes will not start after boot.
If the ipcdir_server is to be restarted for another reason:
1.Log in to the TBS master host as root.
2.Start ipcdir_server by writing:
root@server> /opt/ericsson/nms_cif_tmos_tbs/bin/ ipcdir_server
The following is an example of how to remove logs older than three days:
Table 4 Example Command To Delete All Log Files Older Than Three Days
>./jm_rm_db_log.sh -d 30
./create_sybdev /ossrc/sybdev/oss/sybdata 700 ddev_lvlogdb
################ how to find user log for RNC ###############################
lgolqn <hours>h'
############################ Creating a User Account with the SMC GUI Tool #####
#########################################
Start the SMC by issuing the following command:
/usr/sadm/bin/smc &
################################ how to set date in Solaris ####################
#########################################
The date command takes the syntax mmddhhmmyy, or month day hour minute year . You h
ave to be superuser to set the time.
Example:
# date 1201010106
Fri Dec 1 01:01:00 PST 2006
C- go on /opt/ericsson/sck/lib/create_sybdev
and run the command to increase log and data device
root@htossm1o> ./create_sybdev /ossrc/sybdev/oss/sybdata 700 ddev_lvlogdb
SYB lock successful.
root@htossm1o> ./create_sybdev /ossrc/sybdev/oss/syblog 50 ldev_lvlogdb
SYB lock successful.
D- alter data segment according to the change
sybase@htossm1o> isql -Usa
Password:
1> alter database lvlogdb on ddev_lvlogdb="700M"
2> go
Extending database by 358400 pages (700.0 megabytes) on disk ddev_lvlogdb
1>
************** run the commande below via alom ip to boot the server if it's han
ging for a particular reason ********************
#console -d 0
#type y
#then type ok boot
#for backup on OSS, copy is saved on /ericsson/dmr/log/recovery
Check for the TIME and CPU columns, if CPU column is more than
10% and TIME value is more than 10:00 (10mins), it means that
the process is in question. Get the PID of the process and use /usr/ucb/ps
to check details of process in question
# /usr/ucb/ps -auxww <PID>
******************* check interface VLAN for the backup server
#cat /etc/hosts then type
#netbackup
or connect to another server and run the command below
#ifconfig -a
**************** SMF Commands to enable services *******************************
*************
check faulty services
#svcs -x
svcadm manipulates services:
# svcadm enable enables services, services start when dependencies are rea
dy
# svcadm disable disables services
# svcadm restart stops and starts services
# svcadm refresh commits the current properties to the running snapshot and
runs the refresh method
# svcadm clear signals that a service in maintenance has been fixed
With -s, enable & disable wait until completion
With -t, enable & disable are temporary until next boot
$amosrb_pid = 7241
Last MO: 0. Loaded 0 MOs. Total: 1 MOs.
Preparing offline MIB:
0% ~50%
~100%
Solution:
AMOS should now be working. The problem was that the user was logging their sess
ion using the name/IP of the node in their home directory. This causes a conflic
t with AMOS which attempts to read the file as an IOR file. if there is file whi
ch has the same name as the nodes located in the users home directory, then mosh
ell will consider this file is the IOR file, and it will try to resolve the cont
ents of the file, but actually this is not a real IOR file, it will return the e
rror that this is offline node.
1. Moving file
sgurjeet@htossa1o> pwd
/home/sgurjeet
sgurjeet@htossa1o>
sgurjeet@htossa1o> mv RNCTG01 RNCTG01.orig
sgurjeet@htossa1o> ls -alrt | grep -i rnc
-rw-r--r-- 1 sgurjeet nms 3018 Feb 4 11:46 RNCTG01.orig
2. Starting AMOS
sgurjeet@htossa1o> amos RNCTG01
first way
htossa1o{nmsadm} # echo $DISPLAY
:70.0
htossa1o{nmsadm} # xhost +
access control disabled, clients can connect from any host
htossa1o{nmsadm} # ssh -l root 172.20.216.200
Password:
Last login: Wed Jul 4 17:30:54 2012 from 172.20.216.231
Sun Microsystems Inc. SunOS 5.10 Generic January 2005
You have new mail.
# tcsh
# c
c: Command not found.
#
# cd /usr/openv
# cd java
# setenv DISPLAY 172.20.216.228:70.0
# ./jnbSA &
2nd way
open a terminal
#ssh -X root@172.20.216.200
#jnbSA &
************************ how to set gauge period to 15min for GGSN node ********
*********************************************
root@htossm1o> pwd
/etc/opt/ericsson/ccpdm
root@htossm1o> grep GAUGE_INTERVAL_VAL cxc.env | head -1
GAUGE_INTERVAL_VAL=15
root@htossm1o>
You need to restart PDM processes in order to put it into effect. Please run the
following command in Master Server:
# smtool coldrestart PDM PDM_SNMP reason=other reasontext= change PDM gauge period
Please note it will disturb statistics collection and might result gaps in SNMP
based nodes. So please choose time suitable for you to perform this activity.
al Display alarms
alt Display alarms chronologicaly
ala Display alarms with detail
get UtranCell <parameter> Display table with parameter for all cells
get UtranCell=<cell> <parameter> Display parameter for cell
set cell <parameter> <value> Change parameter to specified value for all cell
s
bl Block
bl Soft Block
deb Deblock
get RncFeature ! Display features enabl
ed on RNC
get RncFeature featureState ! Display features enabled on RN
C
get RncCapacity currentCapacityLimit Display RNC User and Throughput Capacity
get ChannelSwitching Display RNC Channel Switching Parametrs
get Handover Display RNC Handover Parameters
get Paging Display RNC Paging Parameters
get RabHandling Display RNC RAB Handling Parameters
get Hsdsch
get Eul
RBS
get RbsLocalCell Display RBS Parameters
get Iub Disblplay Iub Parameters (maxHsRate)
!Site Availability!
strt
!Mobatch!
moshell -v password=rnc -m RNCARU2,RNCCUR1 'lt all; st utrancell; exit;'
moshell -v password=rnc RNCARU2 'lt ^utrancell; st utrancell; exit;'
mobatch -o /home/swright/ABCSITES.txt 'lt all; alt' /home/swright/mobatch
mobatch -v password=rbs -o /home/swright/ABCSITES.txt 'lt all; lga' /home/swrigh
t/mobatch
!Log of Alarms!
lga
lset BON007X accessClassNBarred 1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0
lset BON007X accessClassNBarred 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
!RBS Resources
ced
ced | sort -k 9 ! Sort on interference
cedh !CE Resource usage
cedh | sort -k 8 !CE ResourceL CEs
Sharing a directory among users in same group is one of the essential tasks. You
need to use chmod command and add user to appropriate group. To make idea clear
here is an scenario:
/home/myproj : is shared directory
usr1, usr2, ... usrN : would like to work and share files in /home/myproj direct
ory
padmin : Main project administrator user
Step # 1: Create a shared directory /home/myproj
If this directory does not exist then create it:
# mkdir /home/myproj
Step # 2: Create the group shared group
You need to create a new group. Let us assume group name is myproj
# groupadd myproj
Step # 3: Add user project administrator (padmin) and setup password:
# useradd -d /home/myproj/ -g myproj -m padmin
# passwd padmin
Step #4: Add rest of users to group myproj
# useradd -d /home/myproj/ -g myproj usr1
# passwd usr1
Add second user:
# useradd -d /home/myproj/ -g myproj usr2
# passwd usr2
... and so on...
Step #5: Setup permission on /home/myproj directory as follows:
(a) Setup group ownership to myproj group:
# chown -R padmin.myproj /home/myproj/
(b) Setup full permission for group and owner on a directory:
# chmod -R 775 /home/myproj/
(c) Setup sgid bit. So what is sgid bit? Normally whenever you creates file in a
directory it belong to default group of user. When a file is created in a direc
tory with the sgid bit set it belogns to the same group as the directory. The re
sult is all users of myproj group can create/alter files in /home/myproj directo
ry:
# chmod -R 2775 /home/myproj/
OR
# chmod -R g+s /home/myproj/
################ HOW TO CHECK PASSWORD ON A node ###############################
#########
how to list nodes and users
root@htossm1o> pwAdmin -list
how to see the password
root@htossm1o> pwAdmin -get SGSNHT02 ossuser
pwAdmin: Error: Invalid option: -get <argument missing>. Use 'pwAdmin -help' for
details.
root@htossm1o> pwAdmin -get SGSNHT02 normal ossuser
FAILED (in server): The given combination of system: 'SGSNHT02 (normal)' and acc
ount: 'ossuser' is not defined in the password service.
root@htossm1o> pwAdmin -get SGSNHT02 NORMAL ossuser
Ericsson@123
root@htossm1o>
how to change password
root@htossm1o> pwAdmin -changePw SOFTWARE NORMAL ossuser -pw<new password>
1) list all nodes info
pwAdmin -list
2) check help to know how to use pwAdmin password
pwAdmin -help
-----------------------------printout:--------------------------------------
Usage: pwAdmin
[-list [ system [type] ] |
-create {system type account [-pw pw] [-info info] | -f file} |
-delete {system type account | -f file} |
-changePw {system type account [-pw pw] | -f file} |
-setInfo {system type account info | -f file} |
-get system type account |
-changeKey [key] |
-init |
-help]
----------------------------------------------------------------------------
3) get current password for the node(example BSCTG10)
pwAdmin -get SYSTEM TYPE ACCOUNT
EX:pwAdmin -get BSCTG10 NORMAL ossuer
ptintout: Pass123
4) set a new password after
pwAdmin -changePw system type account
ex: pwAdmin -changePw BSCTG10 NORMAL ossuser
printout:
Enter new password for ossuser at BSCTG10:
Note1: the password must be 1st defined in node APG by concerned team
note2: after having modified the node password from OSS side, the ONRM database
must be updated by using modify function in ONE.
################################################################################
############################################