[lttng-dev] lttng snapshots and running traces

[Thibault, Daniel]

-----Message d'origine-----
De : Julien Desfossez [mailto:jdesfossez at efficios.com] 
Envoyé : 26 septembre 2013 12:16

> When the consumer needs to read a subbuffer, it performs a get_subbuf operation. If this operation succeeds, 
> the consumer has exclusive access to this subbuffer and so it can read it safely. If it fails, it means the tracer is 
> currently using it and in the case of snapshots, the consumer will just skip to the next one (up to the end 
> position it defined at the beginning of the snapshot).
> So if there are gaps for a stream in the recorded snapshot, their size will be a multiple of the subbuffer size.
>
> Julien

   Ah, so it occurs at the sub-buffer level.  Simple and efficient.

   Is the following scenario possible?  The consumer tries to get_subbuf, and is denied access (because tracers are writing into it).  It then tries to get the next sub-buffer, but the luck of task scheduling is such that by the time it actually calls the request routine, the tracers have already overwritten the sub-buffer in question and gone on to the next one.  The get_subbuf succeeds and the consumer keeps doing its thing (the timestamps remain monotonically increasing, so there is no corruption).  The resulting snapshot on disk has a large jump in its timestamps, having lost a whole buffer cycle's worth of event records.

   To complicate things, the tracers then stop (for lack of event occurrences) in some sub-buffer ahead of the consumer's current position, but still short of the consumer's stop goal.  When the consumer skips over the busy sub-buffer (where the tracers are "stuck"), it would read events that apparently jump back in time (a whole buffer cycle's worth of time), so I guess it would stop there and close the snapshot?

   On a somewhat related topic, how does the session daemon handle the context added to a channel?  Is it a set of instructions (flags, really) sent to the tracer, similar to filter pseudo-code (but much less complicated)?

Daniel U. Thibault
Protection des systèmes et contremesures (PSC) | Systems Protection & Countermeasures (SPC)
Cyber sécurité pour les missions essentielles (CME) | Mission Critical Cyber Security (MCCS)
R & D pour la défense Canada - Valcartier (RDDC Valcartier) | Defence R&D Canada - Valcartier (DRDC Valcartier)
2459 route de la Bravoure
Québec QC  G3J 1X5
CANADA
Vox : (418) 844-4000 x4245
Fax : (418) 844-4538
NAC : 918V QSDJ <http://www.travelgis.com/map.asp?addr=918V%20QSDJ>
Gouvernement du Canada | Government of Canada
<http://www.valcartier.drdc-rddc.gc.ca/>