Rstudio server lm and glm not converging properly (only on datacentre cluster)

Has anyone else encountered a problem with lm and glm not converging in Rstudio server only. The model runs properly when running in R (ssh into the server), on the same version Rstudio server on a linux destop and on windows. Its such weird behaviour that I am at a loss where to look next. Noting I have been using R and Rstudio for a long time and this is the first time I have encountered this.

Note: the Residual deviance blows out and the associated solutions do the same.

Update this one gets even stranger. The problem is now nailed down to when using MKL from rserver. As mentioned before no problems were observer when using r on the server directly.

openblas seems to work without any issues.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.