sanity check on server
We have Linux server with oracle 10g(R2) database in it.My client asked us to perform a sanity check on this server.
since i am new to this work,I have no idea what it is.So,Please someone help me with this thing.
Tell me what and how to do it and how the report should be.
If any one have report of this kind please paste it here,that would be really helpful.
Thanks in advance....
Sanity check could mean a lot.
- any sign of hardware failure? (raid, ram, etc)
- load is ok?
- time and date are right?
- is there enough disc space?
- all logs are rotated?
- do only neccesary daemons run there?
- is the operating system up2date?
- any sign of intrusion?
- all needed links are up, full duplex and gigabit/s (or 100mbit/s) ?
- are interface errors?
- all daemons, that shall be reached externally, can be reached?
- does oracle show any errors?
- is oracle up2date?
- does the size of the database make sense?
I am sure, there are a million things to check in oracle itself, but I am no expert there :)
- does it backup the right data?
- does it backup?
- does a restore work?
- does the backup finish in an appropiate time frame?
A lot of these questions can be answered by using something like xymon, nagios, munin, etc.
Then you can show your client the RRD graphs
and ie say: "Look, no problems in diskspace, etc.
But load is too high for too long, we have to work on a solution for that."