Support can manually generate RAS tickets with either srvclogcli or adminlogcli in Stornext for troubleshooting.

 

##Version qwiki article is being written against

[root@up logs]# cvversions

File System Server:
  Server Revision 4.7.0 Build 36215 Branch branches_4.7.0
  Built for Linux 2.6.18-8.el5 x86_64
  Created on Fri Aug 16 16:57:36 MDT 2013
  Built in /scm/nightly/VM-0-RedHat50AS-26x86-64-SP0/sn/buildinfo
 

 

##Location of binaries

[root@up logs]# file /usr/adic/SRVCLOG/bin/srvclogcli
/usr/adic/SRVCLOG/bin/srvclogcli: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for GNU/Linux 2.6.9, dynamically linked (uses shared libs), not stripped
 

[root@up logs]# file /usr/adic/SRVCLOG/bin/adminlogcli
/usr/adic/SRVCLOG/bin/adminlogcli: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for GNU/Linux 2.6.9, dynamically linked (uses shared libs), not stripped
 

 

You must specify the required parameters when using the *logcli binaries.  Here is the STOUT from the srcvlogcli as an example.

 

[root@up logs]# srvclogcli

Usage:  srvclogcli -f FILE -l LINE -p logPath -r rptFru -j rptFruInst
        -v violFru -i violFruInst -q requestId -e event -t timestamp
        "detailed error text"

Parameters (and defaults):
         -f FILE         file name (= UNKNOWN)
         -l LINE         line number (= 0)
         -p logPath      srvclog location (= /usr/adic/SRVCLOG/logs/srvcLog)
         -r rptFru       reporting FRU, required parameter
         -j rptFruInst   reporting FRU instance (= UNKNOWN)
         -v violFru      violating FRU, required parameter
         -i violFruInst  violating FRU instance (= UNKNOWN)
         -k violParInst  violating Parent FRU instance (= UNKNOWN)
         -q requestId    request ID, optional parameter
         -e event        event identifer, required parameter
         -t timestamp    timestamp in time(2) format (= Now)
         "required detailed error text"

Example: srvclogcli -f myScript -r SL_FRU_SP -v SL_FRU_SNA_FCB_FW
                    -i sna1blade1 -e SL_EVT_WRONG_FW_LEVEL
                    "Wrong firmware level on FC blade: sna1blade1"

                    
                    
                    
        
At minimum you will need to provide the following arguments to generate a RAS manually.

violFru -  This is the violating FRU.  This can be the same as the reporting FRU (rptFru).  This argument exists should one component be monitoring another that fails.
rptFru - This is the reporting FRU.
event - This is the event of the RAS being reported.  

You can reference the voilFru and the rptFru from the mysql database named vdb.

Example:
#To enter the database.
'mysql vdb'

#To only see fru names from the sl_fru_generic table.
'select FRU_NAME from SL_FRU_GENERIC;'

When using srvlog/adminlog cli commands you can make reference the snfs_ras man page to find valid strings that can be used for
this argument. Also the vdb can be reference from the sl_event_generic table.

Example Command:
#The enter the database.
'mysql vdb'

#To view everything in the sl_event_generic table.
select * from sl_event_generic;


Here is an example of the srvclogcli specifying the reporting FRU (-r) as 'SL_FRU_SP' and the violating FRU (-v) SL_EVT_SYS_RES_WARN as a SL_EVT_SYS_RES_WARN event with a message of "Jon testing RAS".
 

Command:

srvclogcli -r SL_FRU_SP -v SL_EVT_SYS_RES_WARN -e SL_EVT_SYS_RES_WARN "Jon Testing RAS"

 

This will generate a RAS ticket in the Stornext MDC GUI as well as send a email event to the recipients set for this level warning.  This can be particularly
helpful if you want to generate a RAS ticket for the customer with out actually inflicting an error.