Ok, that makes sense. I'll have to think of another way to handle this then, maybe I can just run two servers at the same time on different ports for testing purposes. I'll see if that works.
And yeah a UtilityOpProcessor sounds like something that would be very useful and not too hard to setup. I might give it a go. On Tue, Oct 6, 2015 at 3:56 PM, Stephen Mallette <[email protected]> wrote: > > > > Does anyone know if there is a way of changing the gremlin-server > > configuration without a restart. > > > I purposely didn't setup gremlin server that way, as I found in Rexster, > many of the changes that you could make to Rexster's config ended up > basically forcing a restart anyway. It made Rexster's config code very > painful to deal with. I'm sure folks would love to be able to make changes > to Gremlin Server's config dynamically and have it auto-update without > restart, but I think that's a really big blob of work. Not sure what would > be involved with trying to retrofit that in at this point...... > > Also slightly unrelated but is there a way of triggering server errors for > > client side error management integration tests? > > > triggering all the errors? i'm not aware of a manual way to do that. we > could add a new OpProcessor for that though. Like maybe a > UtilityOpProcessor? > > > On Tue, Oct 6, 2015 at 9:08 AM, Dylan Millikin <[email protected]> > wrote: > > > Hey devs, > > > > Does anyone know if there is a way of changing the gremlin-server > > configuration without a restart. Here's my conundrum: > > > > I use Travis CI for testing purposes. I would like to test non-ssl driver > > features and ssl with auth features all in a single run. The single run > is > > important to keep things simple and keep code coverage unified (which I > > wouldn't get if I created multiple build envs) > > > > This requires programmatically changing the server configuration > > (preferably without a restart) and I don't know if this is doable atm. > > Maybe this could be a use to the control opProcessor? > > > > Also slightly unrelated but is there a way of triggering server errors > for > > client side error management integration tests? (asking the server to > > return a specific error and error code). I know that throwing a > > gremlin-groovy script error would only ever generate a 597 error but > maybe > > there's something available here using another op? > > > > Thanks! > > >
