On Thu, 2020-01-16 at 20:18 -0600, Benjamin Marzinski wrote: > There were some variables in the hwe and mpe structs that weren't > being > merged by merge_hwe() and merge_mpe(). > > Signed-off-by: Benjamin Marzinski <[email protected]> > --- > libmultipath/config.c | 8 ++++++++ > 1 file changed, 8 insertions(+) > > diff --git a/libmultipath/config.c b/libmultipath/config.c > index 20e3b8bf..85626e96 100644 > --- a/libmultipath/config.c > +++ b/libmultipath/config.c > @@ -372,6 +372,10 @@ merge_hwe (struct hwentry * dst, struct hwentry > * src) > merge_num(san_path_err_threshold); > merge_num(san_path_err_forget_rate); > merge_num(san_path_err_recovery_time); > + merge_num(marginal_path_err_sample_time); > + merge_num(marginal_path_err_rate_threshold); > + merge_num(marginal_path_err_recheck_gap_time); > + merge_num(marginal_path_double_failed_time);
Hm, looking at this, I think we should not attempt to merge sets of parameters that belong together like this. The marginal_path params should only be merged if they are *all* unset in the destination. Otherwise we risk to end up with an inconsistent parameter set from two different sources. But that's not a failure of your patch, it should be done in a follow- up fix. So: Reviewed-by: Martin Wilck <[email protected]> -- dm-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/dm-devel
