We use coarse grained Langevin dynamics simulations to study packing and ejection of semi-flexible polymers into and out of a spherical capsid inside a crowded cell. The aim is to find the effect of crowding environment on the polymer packing and ejection as a function of crowding particles type and interaction (repulsive versus attractive) with the polymer. We also find in this environment the effect of the capsid tail on the packing process. Such packing and ejection conditions are relevant, for example, to λ DNA packing inside E.coli bacterial cells, where the environment is crowded due to the presence of proteins, bacterial DNA and salts. We use neutral and charged, but highly screened, polymers, and compare packing and ejection rates of the two. For a neutral polymer packing into a capsid with a tail, attractive interactions with the crowd particles make packing slightly harder at higher crowd densities, but repulsive interactions make it easier. Our results indicate that packing into a tailless capsid is less efficient at low crowd densities than into one with a long tail. However, this trend is reversed at higher densities. In addition, packing into a capsid with a long tail shows a highly variable waiting time before packing initiates, a feature absent for a tailless capsid. Electrostatic interactions at physiological conditions do not have much effect. Some bacterial cells, such as Pseudomonas chlororaphis, form a nucleuslike structure encapsulating the phage 201φ2-1 DNA. We also study the packing dynamics with the nucleus present. We find packing is faster compared to the case of no-nucleus packing. We also observe knot formations but these knots untangle vi quickly while the polymer translocates. This knot formation is independent of polymer charge and presence of crowd particles. For ejection process, our data reveal that as the density of the crowded environment increases, the ejection time increases due to entropy loss and the increasing of bead repulsion. However, if these crowded environment attracts the polymer to the cell the ejection process becomes faster because in this environment beads pull the polymer a way from the capsid towards the cell, helping the ejection process. To compare our result with the in vivo experiment in 2012, we plot the velocity of the ejection versus the DNA remaining in the capsid and versus the DNA ejected for the ejection of two polymers differing in length by 27% as done by the experiment. We find that the velocity of the ejection of the longer polymer is faster, which agree with the experimental data. But we did not find any overlap between the two curves.