Jacek, Within the eups framework you have to decide if you want to run from your build directory or an installed directory. The former is more convenient (but will probably require changing the Qserv build process a little bit); the latter is better for provenance and repeatability. You also have to decide if you want total safety (rebuild all configuration and data files on update) or convenience (reuse an existing instance). I think you always have to stop the existing instance and start a new one. But it may be possible to have the new one reuse the configuration and data of the old one. What I think you want is this: # Assume Qserv configuration/instance in ~/run/qserv qserv-stop.sh ~/run/qserv # Assume Qserv source git clone in ~/repo/qserv cd ~/repo/qserv eupspkg -er build eupspkg -er install eupspkg -er decl -C setup qserv qserv-start.sh ~/run/qserv # and now finally I can run the query that causing troubles mysql --protocol TCP --port 4040 -A LSST -e "my query" -- Kian-Tat Lim, LSST Data Management, [log in to unmask] ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the QSERV-L list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1