Commit 468576d1 authored by Thorsten Simons's avatar Thorsten Simons

1.3.1 - documented how to use a different folder for DB tmp indexes

parent 39033ed7
......@@ -61,7 +61,17 @@ Running **hcprequestanalytics** is a two step process:
query results will be falsified, as well.
2. Run queries against the database::
2. Run queries against the database
.. Tip::
**hcprequestanalytics analyze** starts as much subprocesses as CPUs are
available. Using that pool of subprocesses, it runs queries in
parallel. On a 4-CPU system, the overal runtime should go down to
roughtly a quarter; the limiting factors are descibed in the
:doc:`80_goodtoknow` chapter.
::
$ hcprequestanalytics analyze -d hcp72.db -p hcp72
scheduling these queries for analytics using 4 parallel process(es):
......@@ -108,22 +118,6 @@ Running **hcprequestanalytics** is a two step process:
percentile_req : 22.797 seconds
analytics finished after 34.129 seconds
.. Warning::
Some of the provided queries will cause the database to create temporary
indexes, which can use the same capacity on disk as the database itself.
These indexes will go into the default tmp folder configured for the
user, which might not have enough free space to store them. This will
cause the respective queries to fail. If **hcprequestanalytics** prints
error messages about *filesystem or database full*, you can make
sure that an appropriately sized folder is used by setting this
environment variable before running **hcprequestanalytics**::
$ export SQLITE_TMPDIR=/wherever/you/have/enough/space
Make sure to replace ``/wherever/you/have/enough/space`` with a path
that matches your systems reality, of course!
.. Tip::
You can run selected queries by adding them to the end of the command::
......@@ -218,10 +212,3 @@ Running **hcprequestanalytics** is a two step process:
-rw-r--r-- 1 sm staff 1.0K Oct 3 20:36 hcp72-req_httpcode.csv
-rw-r--r-- 1 sm staff 3.4K Oct 3 20:36 hcp72-req_httpcode_node.csv
.. Tip::
**hcprequestanalytics analyze** starts as much subprocesses as CPUs are
available. Using that pool of subprocesses, it runs queries in
parallel. On a 4-CPU system, the overal runtime should go down to
roughtly a quarter; the limiting factors are descibed in the
:doc:`80_goodtoknow` chapter.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment