At times, you might need to verify the system's health (usually) after a hardware replacement.
This article provides the steps to perform a basic Cobalt system health check and verify if the Cobalt server is operating correctly.
Perform the following steps as a smsadm user:
- Verify that the SMSC status is Online:
- Check the start time for all the processes and compare them with the SMSC start time:
ps -ef | grep smsadm
- Verify that all major SMPP clients are in the Connected state (SMPP Client status):
- Verify the number of messages, status, and violations in the current hour (SMPP Server status):
- Ensure that there are no violations:
- Verify the license utilization in SMServer:
- Check the status for all processes:
- Verify all the error codes:
You may need to execute this command several times.
- Check for the important traffic counters from the recent traffic files:
For example, counters like
Delivered SM, and
SS7-MO-Rcvdindicate that the messages are coming to the SMSC, and the SMSC is delivering the messages.
egrep "Time_Stamp|Submit SM-Rcvd|Delivered SM|SS7-MO-Rcvd" TRF.<timestamp>
- Observe and confirm that there is no core file:
find $EBSHOME/* -name "*core.*"
- If you observe a core file, follow the next two steps as a root user.
- Note down the timestamp for the core file by running the following command:
ls -l core.*
- Note down the process that generated the core file by running the following command:
grep -i "exited (FAIL)" $EBSHOME/SMS/$VERSION/RUN/mlog/MLog*
grep -i "exited (KILL)" $EBSHOME/SMS/$VERSION/RUN/mlog/MLog*
(The log in the below command indicates that some process went to the blocked state.)
grep -i "IPC message failed" $EBSHOME/SMS/$VERSION/RUN/mlog/MLog*
(The log in the below command indicates that the shared memory is exhausted.)
grep -i "shared memory capacity exceeded" $EBSHOME/SMS/$VERSION/RUN/mlog/MLog*
grep -i KILL $EBSHOME/SMS/$VERSION/RUN/mlog/MLog* | wc -l
netstat -na | grep 6200