Am 14. Mai 2015 14:14:51 MESZ, schrieb Philippe Mouawad <[email protected]>: >Ok , let's forget about it . >Thanks for feedback.
What about compressing the output on the fly and thus using less space instead? Felix > >On Thu, May 14, 2015 at 2:29 AM, sebb <[email protected]> wrote: > >> On 14 May 2015 at 00:10, Vladimir Sitnikov ><[email protected]> >> wrote: >> >> yes but isn't it then a result of the load test to know that your >db >> > tablespaces sizing or cleanup/archiving was not correct ? >> > >> > True. The problem is humans are prone to errors, so you'll run into >> > that "not correct" issues eventually. >> > >> > Well, precondition check is a common need. >> > I am not sure if it should/could be addressed by JMeter itself. >> >> I agree, this is out of scope for JMeter. >> >> I doubt it can easily be made foolproof, and if it does seem to work >> people will rely on it and then blame us when some unexpected event >> uses up the disk space. >> I don't think it's worth expending any effort on it. >> >> Besides, if the system is so short of disk space that it can run out >> during a test, it's probably already badly fragmented and will >perform >> badly. >> >> >> > The volume required depends on the number of test iterations. >> > Certain our tests include simultaneous execution of multiple JMeter >> > scripts, so it is not that easy for each individual script to >"foresee >> > that there would be other scripts consuming the space". >> > >> > I think "out of space to write jmeter results" is rare issue for >us. >> > >> > Vladimir >>
