Recent years have brought strong observational evidences for the standard LCDM cosmological
model. Cosmic microwave background (CMB) anisotropy and large scale structure (LSS) probes
do not favour any extensions of the standard model. Nevertheless, in this framework, the prefered
cosmological parameters may differ from probe to probe, from experiment to experiment. This is
the well known case of the tension between CMB and Sunyaev Zel’dovich (SZ) galaxy clusters
(GC) from Planck. In 2013, the Planck team has shown that the prefered matter content (WM)
and density fluctuation power spectrum amplitude (s8), the two main cosmological parameters
probed by the galaxy cluster number count, are different in the CMB analyses and in the SZ
cluster analyses at more than 2 sigmas (a result confirmed in subsequent analyses). We present
the results of our new analysis using more recent measurements of the CMB, SZ clusters and SZ
power spectrum of 2016 and show that the tension on (WM,s8) is mostly releaved. The lower
value of the reionisation optical depth and thus of s8 in the recent Planck studies is the main
reason. We also show that basic extensions of the standard model (massive neutrinos or non–
lambda dark energy) do not help improving the agreement between the probes. In order to fully
reconcile SZ clusters with CMB best model, the mass of the galaxy clusters should be 40% lower
than derived from hydrostatic equilibrium estimates. While current numerical simulations and
weak lensing measurements agree for a mass bias of 20%, investigations are still going on to
explain such disagreement on the mass bias. We show that considering a mass bias evolving with
redshift or mass does not help in eliminating the discrepancy.