I tried the former 0.13 version to check if that is a regression.
The format changed a bit and some support was not there.
So let us not think about the latter 6 events (those need a newer feature
which isn't in 0.13).
But the first 4 events can be found and are only present in some runs.
# for i in fatrace.log.*; do ./grepcheck $i; done
fatrace.log.4Y4sj : ....FFFFFF - 6 matches missing
fatrace.log.BmDhE : ....FFFFFF - 6 matches missing
fatrace.log.D3Q0R : .FF.FFFFFF - 8 matches missing
fatrace.log.dBKeg : ....FFFFFF - 6 matches missing
fatrace.log.DwbX5 : ....FFFFFF - 6 matches missing
fatrace.log.FfiSG : .FF.FFFFFF - 8 matches missing
fatrace.log.HZY4w : ....FFFFFF - 6 matches missing
fatrace.log.KQgf3 : ....FFFFFF - 6 matches missing
fatrace.log.kWx9y : ....FFFFFF - 6 matches missing
fatrace.log.lM669 : FFFFFFFFFF - 10 matches missing
fatrace.log.mYvpw : ....FFFFFF - 6 matches missing
fatrace.log.n78BL : ....FFFFFF - 6 matches missing
fatrace.log.qr7ll : ....FFFFFF - 6 matches missing
fatrace.log.rglRA : ....FFFFFF - 6 matches missing
fatrace.log.s3gPp : ....FFFFFF - 6 matches missing
fatrace.log.sktJy : ....FFFFFF - 6 matches missing
This leads to the assumption that this never worked better, but now is
catched by the tests added in 0.15.
How can this test well in Debian CI as I can reproduce in Debian just as
well?
The answer - just as Brian pointed out initially - still seems to be:
checking log...
skipping FAN_REPORT_FID tests, not available on this kernel
It will most likely hit Debian once the kernel in the test env gets bumped.
Since (so far) only the --currentmount tests failed I tried if it is
different without that.
It is, but only statistically - I get ~50% good results then - maybe a
timing/buffer issue that works better but not perfect without this?
# while /bin/true; do f=$(mktemp /tmp/fatrace/fatrace.log.XXXXX); mv
/tmp/fatrace/fatrace.log $f; ./tests/fatrace-currentmount; done
# for i in fatrace.log.*; do ./grepcheck $i; done
fatrace.log.0xLev : .......F.. - 1 matches missing
fatrace.log.2Jmxh : ........FF - 2 matches missing
fatrace.log.3USXR : ......F... - 1 matches missing
fatrace.log.44Di7 : .FF.FF.... - 4 matches missing
fatrace.log.83JEz : ...F...... - 1 matches missing
fatrace.log.9cd6V : .......... - 0 matches missing
fatrace.log.9U2XK : .FF..F.... - 3 matches missing
fatrace.log.a5lNX : .......F.. - 1 matches missing
fatrace.log.bsAZe : ......FF.. - 2 matches missing
fatrace.log.bZxDG : .......... - 0 matches missing
fatrace.log.CN2Gc : .......... - 0 matches missing
fatrace.log.daVJ2 : .......... - 0 matches missing
fatrace.log.DrQVf : .......... - 0 matches missing
fatrace.log.eT5r1 : ......F.FF - 3 matches missing
fatrace.log.F0P6T : .......... - 0 matches missing
fatrace.log.f4Ak7 : .......... - 0 matches missing
fatrace.log.H2oRi : ....F..F.. - 2 matches missing
fatrace.log.inSK6 : ...FF..F.. - 3 matches missing
fatrace.log.iSJJG : ......F.FF - 3 matches missing
fatrace.log.Jqxrk : ...F...... - 1 matches missing
fatrace.log.L912a : ......F... - 1 matches missing
fatrace.log.n5aUL : .......F.. - 1 matches missing
fatrace.log.Npaef : .......... - 0 matches missing
fatrace.log.updMK : .......... - 0 matches missing
fatrace.log.WVFtI : .......... - 0 matches missing
fatrace.log.x50aO : ......FF.. - 2 matches missing
fatrace.log.XnqDl : .......... - 0 matches missing
fatrace.log.yPsnf : ......F... - 1 matches missing
fatrace.log.zBbGh : .......... - 0 matches missing
Still makes no sense to me - is this generally broken or am I missing
something?
--
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
