eG Help

Measures reported by HanaVFileSizeTest

Since the SAP HANA database resides in volatile main memory, a persistency layer in the SAP HANA database engine ensures the durability of the transactions. In the event of a failure the persistency layer ensures that the database can be restored to the most recent committed state after a restart. Transactions are either completely executed or completely un-done. To ensure durability of data changes completed in the memory of the SAP HANA server nodes, database data is written to disks. The persistency layer stores data in disk volumes, organized in pages. There are two types of volumes: data and log. Every process/service in the SAP HANA database which needs to persist data, such as the name, index and statistic servers, writes regularly into its own data and log volume. The data volume contains data pages and undo log information. The log volume only records redo log information. In the event of a crash or recovery, redo log information is used to roll forward changes that have not been persisted on the data pages while undo log information is used to roll back changes for uncommitted transactions. If the data or log volume mapped to a service does not have sufficient free space, then these volumes will not be able to store critical undo/redo log details and data pages, thus causing the disaster recovery engine of the database server to fail! This in turn may cause critical data loss and service failures! To prevent such irredeemable losses, it is of utmost importance to closely monitor the space usage in the data and log volumes of each service, detect a space crunch even before it occurs, and right-size the volumes accordingly. To help administrators with this, the eG agent runs the HanaVFileSizeTest test on the SAP HANA database server.

For each service of the SAP HANA database server, this test reports how that service uses each of its disk volumes (data or log), thereby pointing administrators to services that are over-utilizing their volumes and may require additional space if any permanent damage to performance is to be averted.

The measures made by this test are as follows:

Measurement	Description	Measurement Unit	Interpretation
totalUsedSizeMB	Indicates the amount of space that this service is currently using in this volume.	MB	A low value is desired for this measure.
usedPercent	Indicates the percentage of space that this service is currently using in this volume.	Percent	A low value is desired for this measure. If a consistent growth is noticed in the value of this measure, it is indicative of gradual erosion of space in a disk volume. You may want to free up more space by either adding more disks or by removing unnecessary or obsolete files from the disk. This depends upon the volume type. For instance, in case of the DATA volume, you must not delete files from this volume manually, as it can corrupt the SAP HANA system. On the other hand, if the TRACE volume runs out of space, you can delete the trace files to make room for more trace data.
growth	Indicates the rate at which this volume file has grown since the last measurement period.	MB/Sec	If a consistent growth is noticed, it is indicative of gradual erosion of space in a disk volume. You may want to free up more space by either adding more disks or by removing unnecessary or obsolete files from the disk. This depends upon the volume type. For instance, in case of the DATA volume, you must not delete files from this volume manually, as it can corrupt the SAP HANA system. On the other hand, if the TRACE volume runs out of space, you can delete the trace files to make room for more trace data.