Stress-testing sysmobts of osmo-bts

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/OpenBSC@lists.osmocom.org/.

Harald Welte laforge at gnumonks.org
Mon May 6 06:40:04 UTC 2013


Hi Holger,

On Sat, May 04, 2013 at 09:23:31PM +0200, Holger Hans Peter Freyther wrote:
> On Sat, May 04, 2013 at 05:16:54PM +0200, Holger Hans Peter Freyther wrote:
> > What is not done:
> 
> > * Create a talloc pool for the primitives (or re-use the msgb across multiple
> > reads).
> 
> * The stress test is either showing memory fragmentation or a silent leak.
> The virtual size of the address space is increasing over time, the talloc
> report does not show an increase in memory usage and I switched off the bts
> before I could look at vmRSS on the 'status' file.

does this only happen with your new patches/branch, or does it already
happen in current master?

Looking at your patches, I don't see anything wrong with how msgb's are
allocated and free'd

-- 
- Harald Welte <laforge at gnumonks.org>           http://laforge.gnumonks.org/
============================================================================
"Privacy in residential applications is a desirable marketing option."
                                                  (ETSI EN 300 175-7 Ch. A6)




More information about the OpenBSC mailing list