[OpenAFS] Tracing VLDB queries
Robert Milkowski
milek@wp.pl
Mon, 3 Dec 2018 15:02:55 -0000
>Once trace points are added tracing can be used to answer all sorts of =
questions. =20
Yes. We've been tracing such information (and others) using DTrace/PID =
provider for years, however having those more abstract probes is going =
to make it easier for everyone,
especially in cases where you are not debugging a bug in AFS itself but =
rather debugging some peculiar client(s) behaviour, etc.
>The specific output from your example is very similar to the
>data that is collected by the baked in audit infrastructure. The
>following is output from AuriStorFS vlserver:
[...]
>In OpenAFS the audit infrastructure can be enabled per-service and its
>output can be set to files, named pipes, syslog, Linux message queues
>and on AIX its integrated with the OS auditing system.
Yes, for this specific example you are right.
However the new probes allow you to easily answer many other questions, =
like what's the distribution of how long queries take (and then per =
client, per volume, etc.),
or correlate queries to I/O to disks or network traffic, or CPU, etc.