Exadata FAQ's

Exadata FAQ's

QUESTIONS AND ANSWERS

Q: How can I tell the current cell software version and also available versions to rollback to?
A: Useful commands
• imagehistory
• imageinfo

Q: I’ve heard I should keep an eye on the temperature of the cells. How can I do that?
A: Inlet Ambient Temperature across all cells can be determined with the following command:
dcli -g cell_group -l root "ipmitool sensor | grep 'Inlet Amb Temp'"

Q: I see STORAGE in my explain plan but it doesn't appear that my query is offloading based on runtime and IO throughput rates. Why?
A: This can happen if the query is run serially and direct reads aren't chosen.
• Offloading only happens for direct reads (reads to PGA)
• PQ processes generally always use direct reads,therefore offload
• Foreground processes running serially may choose direct or buffered reads
• Note that in some cases buffered reads are actually better

Q: My offload performance was great, but all of a sudden it degraded by over 20%. What may be causing this?
A: Check for a cell reboot and resultant offline disks. Also, check for issues above the storage layer such as a change in execution plan, which can cause this.

Q: One of my cells rebooted. What should I do?
A: Confirm if the cell is back up and if griddisks are online in asm. Check to make sure each cell has 34 griddisks online:
SQL> select failgroup, mode_status, count(*) from v$asm_disk group by failgroup, mode_status;
Also consider these sources for information:
• ASMCMD> lsdg (see offline_disks column in output)
• ASM instance alert log
• v$asm_disk.repair_timer (seconds remaining until drop)
• v$asm_disk.mode_status

Q: My ASM instance is hung trying to discover disks?
A: Confirm hung on discovery (instead of hang for a different reason) via ASM instance alert log. Once confirmed, try kfod to see if it hangs as well.
• If kfod hangs, check that libskgxp11.so has the same size and checksum as libskgxpr.so in all ORACLE_HOMEs (includingCRS_HOME). If not, relink the oracle binary with RDS.
• Check cellinit.ora and cellip.ora files for correctness
• Check both master and slave diskmon trace files

Q: I suspect network problems. How do I know my InfiniBand fabric is okay?
A: Refer to MOS note: 1053498.1 & try the following utilities:
• These return fabric-wide info, so run from one node, not all
• /usr/bin/ibdiagnet –c 1000
• Check for errors/warnings summary at the end of output
• Search for details in log files (/tmp/ibdiagnet*)
• /opt/oracle.cellos/SupportTools/ibdiagtools/infinicheck
• This script generates a lot of IB traffic to measure throughput,
so it will impact performance while it is running
• /usr/sbin/iblinkinfo.pl Shows link status on all ports

Q: How to cache an object in smart cache?
A:DBA can cache an object by setting CELL_FLASH_CACHE setting to KEEP. The default value of this setting is DEFAULT.

Q: What are Exadata Cell node Unique Features?
A: Cell node unique feature:
Smart Scans
Hybrid Columnar Compression
Storage Indexes
Flash Cache
ExadataI/O Resource Management in Multi-Database Environment

Q: What is the location of Cell Nodes Logs and Traces?
A: Cell Nodes Logs and Traces
$ADR_BASE/diag/asm/cell/`hostname`/trace/alert.log
$ADR_BASE/diag/asm/cell/`hostname`/trace/ms-odl.*
$ADR_BASE/diag/asm/cell/`hostname`/trace/svtrc__0.trc -- ps -ef | grep "cellsrv 100"
$ADR_BASE/diag/asm/cell/`hostname`/incident/*

/var/log/messages*, dmesg
/var/log/sa/*
/var/log/cellos/*

Q: How to disable to Smart Scan?
A: Cell_offload_processing=false (default true)

Q: How to disable FPLIB?
A: _kcfis_cell_passthru_enabled=true (default false)

Q: How to disable storage index?
A: _kcfis_storageidx_disabled=true (default false)

Q: How to disable flash cache issue?
A: For 11.2.0.2, _kcfis_keep_in_cellfc_enabled=false (default true)
For 11.2.0.1, _kcfis_control1=1 (default 0)

Q: General Cell related Database views?
A: Cell related Database view's:
select * from sys.GV_$CELL_STATE;
select * from sys.GV_$CELL;
select * from sys.GV_$CELL_THREAD_HISTORY;
select * from sys.GV_$CELL_REQUEST_TOTALS;
select * from sys.GV_$CELL_CONFIG;

Q: What is Bloom Filter in Exadata?
A: Bloom filter in Exadata:
In Oracle 10g concept of bloom filtering was introduced.When two tables are joined via a hash join, the first table (typically the smaller table) is scanned and the rows that satisfy the ‘where’ clause predicates (for that table) are used to create a hash table.During the hash table creation a bit vector or bloom filter is also created based on the join column.The bit vector is then sent as an additional predicate to the second table scan.After the ‘where’ clause predicates have been applied to the second table scan, the resulting rows will have their join column hashed and it will be compared to values in the bit vector.If a match is found in the bit vector that row will be sent to the hash join. If no match is found then the row will be disregarded.
On Exadata the bloom filter or bit vector is passed as an additional predicate so it will be overloaded to the storage cells making bloom filtering very efficient.

Q: How to Identify a Bloom Filter in an Execution plan:
A; You can identify a bloom filter in a plan when you see :BF0000 in the Name column of the execution plan.

Q: How to disable Bloom Fliter?
A: To disable the feature, the initialization parameter _bloom_pruning_enabled must be set to FALSE.

Comments